Method to compensate trembling, method to control compensation of trembling and image capture device, and method of their control

FIELD: instrumentation.

SUBSTANCE: invention relates to methodology of image trembling compensation as a result of hands trembling, etc. to prevent image deterioration. The device comprises a unit of trembling compensation by movement on a parallel optical axis, a unit exciting the compensation trembling unit, a unit of trembling compensation unit position detection, a unit of compensation value calculation.

EFFECT: performance of compensation for reciprocal trembling without addition of a sensor for detection of reciprocal trembling.

14 cl, 31 dwg

 

The prior art INVENTIONS

The technical field to which the invention relates

[0001] the Present invention relates to a method of compensating image vibration caused by hand shake, etc., to prevent deterioration of the image.

Description of the prior art

[0002] in Order to prevent the trembling of image due to hand shake, etc., issued by the camera, which contain a vibrating device that uses the block compensation shake, exciting unit and the detection unit vibration and so on, and there are fewer factors leading to the failure of the photographer when photographing.

[0003] will Now be briefly described device for compensation of jitter. Detect camera shake caused by hand shake, etc., and control the movement of the lens to compensate for image vibration (hereafter referred to as "compensation lens"), or an element of image formation is carried out depending on the values of the detection, so that the photographing without any image vibration can be performed, even when shaking hands in the operation of photographing. In this case it is necessary to accurately detect the vibration, to compensate for changes in the optical axis due to vibration. In principle, when vibration is detected, the detection unit vibration issue is play the calculation process according to the detection result, such as acceleration, angular acceleration, angular velocity, angular displacement, etc., flickering of the image is suppressed by the control unit to compensate the jitter based on the results of this calculation process, to encourage the compensation lens or the imager to the movement.

[0004] the Device for compensation of jitter, which detects the rotational jitter among shake hands using measuring angular velocity for initiation of the compensation lens or element of the formation of the image to reduce image vibration, installed on a variety of optical devices. However, photographing at a small distance (photographing condition to the strong increase of photographing) accompanied by vibration, which cannot be detected by only measuring the angular velocity. That is, a so-called progressive tremor, applied in horizontal or vertical direction in the plane perpendicular to the optical axis of the camera, and the deterioration of the image due to this also cannot be ignored. For example, in the case of macroporosity, in which the photographic object is approaching nearly 20 cm, it is necessary to directly detect and compensate for the translational shake. In addition, even when photographing a photographic object, RAS is than necessary, at a distance of about 1 m, it is necessary to detect and compensate for the translational shake in the condition in which the focal length of the optical system forming the image is quite large (e.g., 400 mm).

[0005] Laid the Japan patent No. Hei7-225405 disclose the methodology for calculating the translational shake of the integral of the second order acceleration detected by the accelerometer to initiate the compensation unit jitter, together with the output provided another measure of angular velocity. The output of the accelerometer used to detect translational shake susceptible to environmental changes, such as noise, temperature change, or so on, So these destabilizing factors are additionally strengthened by the integration of the second order, and it is difficult to compensate for the translational shake with high accuracy.

[0006] Laid the Japan patent No. 2010-25962 discloses the calculation of the translational shake under the assumption that it is considered as the angular jitter, when the center of rotation is located at a location remote from the camera. In this way feature measuring angular rate and accelerometer; calculated offset value and the angle from the radius of rotation of the rotational shake of their findings for compensation of jitter. Through calculating, key writing, the center of rotation only in the frequency bands immune to interference, it is possible to eliminate the loss of precision due to the destabilizing factors of the accelerometer, as mentioned above.

[0007] In the prior art have the following situation in compensation of translational shake.

[0008] In the method for which the accelerometer is used as detection means for performing compensation of translational shake, it can lead to an increase in the size and cost of the camera. In addition, although it is desirable that the position of attachment of the accelerometer was the position of the principal point of the lens, it is difficult to set the accelerometer near the position of the principal point of the lens.

[0009] in Addition, laid the Japan patent No. 2010-25962 discloses a means of detecting the jitter of the output means of forming the image instead of the accelerometer as a means of detecting shake. In case of detection of jitter from the output means of forming the image there is a way to calculate the compensation coefficient in accordance with a ratio between jitters and the rotational shake immediately before the photographing operation, and multiplying the rotational jitter by a factor compensation in the photographing operation. In this case, the compensation of translational shake can only be performed in the photographing operation. Also, when management compensation shake clip electronic image, the compensation coefficient is calculated when photographing a moving image in accordance with a ratio between jitters and the rotational shake and the amount of translational shake is calculated by multiplying the rotational jitter by a factor compensation. Although it is possible to change the position of the clipping image captured by the imager, depending on the magnitude of the translational shake, there is a situation in which the angle of view becomes unprofitable narrow in the cut-off part of the image, and so on

[0010] in Addition, laid the Japan patent No. 2010-25962 also discloses a means of detecting acceleration of the translational shake of the current flowing through the field coil, instead of the accelerometer as a means of detecting shake. However, this management tool compensation jitter cannot be performed immediately before the photographing operation. If the effect of translational shake significant, for example, when photographing macroregion, it may be difficult to specify the detailed composition and focus accurately, and compensation of translational shake cannot be performed when photographing a moving image. In addition, the Kolka characteristics of the mechanism of compensation of jitter are not considered when evaluating the acceleration values of the current coil, there is a concern that an accurate assessment of acceleration can be difficult, and the accuracy of the estimates may be disadvantageously reduced due to the use of estimated acceleration to compensate for the shake.

[0011] Therefore, the aim of the present invention is to provide a device for compensation of jitter and the capture device image, compact and very mobile, and will allow very precise compensate for image vibration in the translational shake even in a state of photographing by changing the position of the device.

The INVENTION

[0012] an Aspect of the present invention is a device for compensation of jitter that is configured to compensate for jitter by using the excitation to compensate the jitter, and the device includes a compensation unit jitter configured to compensate for jitter by moving not parallel to the optical axis excitation unit configured to excite the compensation unit jitter signal-based guidance excitation, detection unit position, is configured to detect the position of the compensation unit shake to output a signal detection position, and the block calculating compensation, configured to detect a component of the translational shake in shake device of the moving block is and avoid the shakes, so the signal specifying the excitation and signal detection positions are used as input, and calculating compensation unit to compensate the jitter of the magnitude of the jitter of the device that includes the component of the translational shake, to output it as a signal indication of excitement.

[0013] In accordance with the present invention it is possible to perform highly accurate compensation for the translational shake without adding a sensor to detect the translational shake even able to photograph that accompanies the change of position compact and very mobile device.

[0014] Additional features of the present invention will become apparent from the following description of exemplary embodiments (with reference to the attached drawings).

BRIEF DESCRIPTION of DRAWINGS

[0015] Fig. 1A is a diagram showing the direction of movement capture devices.

[0016] Fig. 1B is a diagram schematically showing the imager equipped with a device for compensation of jitter in accordance with the first embodiment of the present invention.

[0017] Fig. 2 is a exploded perspective view showing the configuration example to illustrate the mechanism of compensation shake together with Fig. 3.

[0018] Fig. 3 is a diagram showing the mechanism of whom is Ansatie shake, if you look in the direction of the optical axis.

[0019] Fig. 4 is a block diagram showing an example configuration of the control unit feedback compensation mechanism shake.

[0020] Fig. 5A is a diagram showing the vibration excited by a unit mechanism for compensation of jitter through simulation.

[0021] Fig. 5B is a diagram showing a model of the vibrations with one degree of freedom.

[0022] Fig. 6 is a block diagram showing an example of the configuration of the device to compensate for jitter in accordance with the first embodiment of the present invention.

[0023] Fig. 7 is a block diagram showing a configuration example of device evaluation in Fig. 6.

[0024] Fig. 8 is a diagram illustrating the center of rotation jitter attached to the capture device image.

[0025] Fig. 9A and 9B is a flowchart of sequences of operations illustrating an example of device operation compensating jitter in accordance with the first embodiment of the present invention.

[0026] Fig. 10 is a block diagram showing an example of the configuration of the device to compensate for jitter in accordance with the second embodiment of the present invention.

[0027] Fig. 11A and 11B is a timing diagram illustrating the calculation of compensation of translational shake in accordance with a second embodiment of the present invention.

[0028] Fig. 12 - unit-CX is mA showing an example of a device configuration of the device to compensate for jitter in accordance with the third embodiment of the present invention.

[0029] Fig. 13A and 13B is a diagram illustrating the process unit to determine the status of the oscillations in accordance with a third embodiment of the present invention.

[0030] Fig. 14 is a block diagram showing an example of the device configuration of devices compensating jitter in accordance with the fourth embodiment of the present invention.

[0031] Fig. 15 is a block diagram showing an example of the device configuration of devices compensating jitter in accordance with the fifth embodiment of the present invention.

[0032] Fig. 16 is a block diagram showing an example of the configuration of the device to compensate for jitter in accordance with the sixth embodiment of the present invention.

[0033] Fig. 17 is a diagram to illustrate the management compensation of jitter in accordance with the sixth embodiment of the present invention.

[0034] Fig. 18 is a block diagram showing a configuration example of device evaluation in Fig. 6.

[0035] Fig. 19A and 19B is a diagram illustrating the definition of pan and assessing the speed of parallel motion.

[0036] Fig. 20A and 20B is a flowchart of sequences of operations illustrating an example of the work device to compensate for jitter in accordance with the seventh embodiment of the present invention.

[0037] Fig. 21 is a block diagram showing an example of the device configuration of devices compensating jitter in accordance with the eighth embodiment of the present invention.

[0038] Fig. 22 is a block diagram showing an example of the device configuration of devices compensating jitter in accordance with the ninth embodiment of the present invention.

[0039] Fig. 23 is a block diagram showing a configuration example of device evaluation to illustrate the device to compensate for jitter in accordance with the tenth embodiment of the present invention in conjunction with Fig. 24.

[0040] Fig. 24 is a diagram illustrating the estimated speed of the parallel motion.

DESCRIPTION of embodiments

[0041] the capture Device image in accordance with each embodiment of the present invention is described below in accordance with the enclosed drawings. The present invention can be applied not only to digital odnolistovoy SLR and digital compact camera, but also to a device for photographing, for example, digital video camera, viewing a television camera, webcam or mobile phone.

[0042] Fig. 1A is a diagram representing a direction of shake of the device 101 of the image capture. Device for compensation of jitter performs compensate the shake to shake, caused in connection with the rotation of the device, centered on an axis perpendicular to the optical axis 102 of the optical system forming an image (hereinafter called the "rotational shake", see arrow 103p, 103y). Device for compensation of jitter also performs compensation of jitter in the jitter caused in connection with the rotation of the device, centered at the jitter caused in connection with the translational movement of the device in the direction perpendicular to the optical axis 102 (hereinafter called "translational shake", see arrow 104p, 104y). In addition, for the three-dimensional coordinates X, Y, Z, shown in Fig. 1A, the direction of the Z axis coincides with the optical axis, and two axes perpendicular to it are the X and y Direction around the X-axis is a direction of rotation around the horizontal axis (see arrow 103p), and the direction around the Y-axis is a direction of rotation around the vertical axis (see arrow 103y). The direction of translational shake indicated by the arrow 104y, parallel to the X-axis and the direction of translational shake indicated by the arrow 104p, parallel to the y-axis.

[0043] [First variant implementation]

[0044] the Following describes the first version of the implementation of the present invention.

[0045] Fig. 1B is a top view, schematically showing the device 101 of the image capture device equipped with the your compensation of jitter in accordance with the first embodiment. The block configuration of the image-forming device 101 of the image capture and functional unit of the compensation process image vibration performed in the CPU 106 (CPU) shown in Fig. 1B.

[0046] the Compensation lens unit 111 compensate for jitter and element 107 of image formation are arranged on the optical axis 102 of the optical system forming the image. Element 107 imaging converts the image of the photographic object formed by the optical system forming the image, in the image. The meter 108 angular velocity is a means of detecting the angular velocity to detect rotational vibration, and the detection signal of the angular velocity is displayed in the unit of calculating compensation. Unit 106A calculating compensation for rotational shake is the first means of calculating compensation for calculation of compensation shake the image relative to the rotational shake. In addition, the block 106B calculating compensation of translational shake is the second means of calculating compensation for calculation of compensation shake the image relative to the translational shake. The output of these blocks calculate the value of compensation is transferred vozbujdayuschego nl the ku 110 after addition. Exciting unit 110 excites the compensation lens unit 111 compensate for jitter and compensates the jitter, taking into account the angular and translational shake.

[0047] In the conventional device, a physical sensor such as an acceleration sensor, is provided for detecting the translational shake indicated by the arrows 104p, 104y in Fig. 1A, and the detection signal is sent to the unit 106B calculating compensation of translational shake. In contrast, in this embodiment, the detection of translational shake is performed using a signal derived from the RF block 110 to block 106B calculating compensation of translational shake in place of a physical sensor such as the acceleration sensor for detecting the translational shake. In addition, the details will be described below.

[0048] In the example shown in Fig. 1B, the compensation lens unit 111 compensation shake moves in the plane perpendicular to the optical axis based on the calculated amount of the compensation, but can also be used a method of performing compensation shake by moving the imager in the plane perpendicular to the optical axis. In addition, there is a way to reduce the effect of camera shake by electronic compensation drug is ment by changing the position of the clipping image in each photographic frame, the extracted element, imaging, and compensation of the image vibration can also be performed in combination with many ways to avoid the shakes.

[0049] the following describes an example of the configuration of the block 111 compensation shake with reference to the exploded perspective view in Fig. 2.

[0050] the Base block 401 111 compensation shake holds together the shutter mechanism and the mechanism of the neutral density filter. The base 401 is supplied finger pin 402 as an integral part, and includes a movable finger pin (not illustrated). Three curved grooves are created on the cylindrical Cam (not illustrated) on the outside in the radial direction of the base 401. Finger pin 402 is inserted into the curved groove and moves forward and backward along a curved groove in the direction of the optical axis, but the details are omitted.

[0051] the Group 406 of the compensation lenses entirely held by the holder 416 lens shift using a nail with a seal (not illustrated). The casing 403 lens has an aperture that limits the power of the optical radiation passing through the group 406 of the compensation lens, and an aperture 405 is created on each of the three arms 404 provided on the sides. The protrusions 415 provided in three places are created on the sides of the holder 416 linsys shift, and by planting these protrusions respectively into the holes 405 casing 403 lenses entirely held by the holder 416 lens shift. The magnets 412, 413, which constitute the electromagnetic mechanism, are held by the holder 416 lens shift.

[0052] the Holder 416 lens shift using the welding pressure is attached to the base 401 by three balls of 407. That is, each ball 407 is movable supporting element relative to the holder 416 lens shift. Holder 416 lens shift freely movably supported relative to the base 401 in the direction perpendicular to the optical axis, with the rolling of each ball 407. In accordance with this configuration, since the results in the possibility of obtaining a lower amplitude, the greater frequency of vibration can be obtained in comparison with the procedure using the guide bar, and the best compensation image vibration will be feasible in a digital camera, which includes increasing the number of pixels.

[0053] the Thrust of the spring 414 is a tool offset to offset holder 416 lens with a shift in the direction closer to the base 401. The thrust of the spring 414 is spring tension, which on one side engages with the nail with a beard holder 416 lens shift, and the other side coupled with GW is stem with a beard (not illustrated), created on the base 401. In addition, the radial spring 417, 418 are the tool offset is provided to prevent rotation of the holder 416 lens shift, and attach with nails with a beard (not illustrated) generated respectively on the holder 416 lens-shift and base 401.

[0054] the Coil 408, 409 are held resinous frames 410, 411, respectively. The metal plug is fully implemented on top of the frame, and connect the end of each coil. By prepaymania conductive pattern of the flexible printed circuit Board 424 (hereinafter abbreviated as "FPC") to this metal connector power is supplied from circuit parts to each coil. To provide power to the coils 408, 409, coil 408, 409 is electrically connected with the FPC 424 at site 425 by means of a metal plug. Elements 422, 423 on the Hall effect are used as a means of detecting the position and are located close to the magnets 412, 413, respectively, to detect changes in the magnetic field due to the movement of the magnets. The amount of movement of the holder 416 lens shift is calculated on the basis of the signal of the magnetic detection. Elements 422, 423 on the Hall effect are also implemented in FPC 424, and receiving power.

[0055] FPC 426 is an element wiring to provide power vozbujdayuschego the shutter unit and estrelinha filter and is attached to the holder 420 FPC with FPC 424. The FPC holder 420 is provided with a cylindrical protrusion 421 and is fixed after it is pressed holes FPC 424 and 426 to perform their positioning.

[0056] Fig. 3 is a front view of the block 111 compensation from the photographic object.

[0057] the Notches 428 are receiving portions located in positions three balls 407, located near the compensation lenses, respectively, that is, the vertices of the triangle. Each of the balls 407 is made in three grooves 428, created on the base 401, and each ball is attached by pressure welding to the bracket 416 lens shift through the point of contact. This configuration enables control with low friction, whereby the compensation lens encourage to follow the target position with high precision, and can improve the accuracy of estimation of the translational shake.

[0058] the following describes a method of control of the compensation lens.

[0059] Fig. 4 is a block diagram illustrating the process of calculating the driving unit 110.

[0060] the Target group's position 406 of the compensation lens is entered in block 601 subtraction. Block 601 subtraction subtracts the current position indicated by the detection signal of the position from the target position to calculate the deviation. Signal detection position is a digital signal generated by an analog-citrofortunella output value of the detection element position, such as the element on the Hall effect. At this moment the noise of the observations (see references Zeta) is included in the output value of the item on the Hall effect, etc., and it is served in block 605 addition. Noise monitoring is a component of the noise, the effect of the noise present in the sensor, and electric induction noise outside, etc. and component noise monitoring will be applied to the actual position of the compensation lens as an output element on the Hall effect.

[0061] the Deviation calculated by the block 601 subtraction is output to the controller 602 of the feedback control (see C(s) Fig. 4), and the controller performs control to bring the deviation to zero, namely, that the detected sensor position followed to the target position. However, after the noise component of the system (see the designation (d) applied to the output of the controller 602 of the feedback control unit 603 addition, it appears in the mechanism of compensation of jitter, which is installing 604 (see P(s) Fig. 4), and it will be applied exciting force. The noise of the system is mainly perturbing force due to the influence of gravitational acceleration caused by changing the position of the device image capturing, and the influence of the vibration caused by hand shake, etc., the Compensation lens which is fur the lowland compensation shake, is excited in accordance with the target position and the characteristics of the controller 602 of the feedback control, installation 604, optionally subjected to the influence of system noise and noise monitoring.

[0062] the following describes a method for detecting the translational shake, using the mechanism of compensation of jitter and its a stimulant.

[0063] First, a mechanism for compensation of jitter is simulated setup for two perpendicular to each other axes, as shown in Fig. 5A. Each of the orthogonal axes A and B represents the axis of excitation movable lens (back lens). Excited by block 701 mechanism for compensation of jitter have inertial mass m and is excited for each axis by the driving unit. In connection with the first exciting unit spring constant is represented in the form kathe attenuation coefficient in the form of caand effort in the form of fa; in connection with the second stimulating unit spring constant is represented in the form kbthe attenuation coefficient in the form of cband effort in the form of fb.

[0064] Fig. 5B illustrates a model of the vibration of one axis of the above two axes. That is, it is a schematic view in which the vibration is modeled with one degree of freedom excited for block mechanism for compensation of jitter, including whom ensational lens. Unlike excited by block 701 compensation mechanism shake the mounting block 801 refers to the part which supports excited by a unit (the camera body). In this model, the absolute displacement excited by block 701, which includes a compensation lens, denoted zband the absolute displacement of the fixing unit 801 is denoted zw. In addition, in connection with the agitation unit spring constant is denoted by k, the attenuation coefficient is denoted by c, and the force caused by the power supply to the coil of the compensating mechanism shake, denoted f. The equation of motion of a system model for induced block 701 is shown by the following formula (1) (mass m) in the notation of Newton:

[0065] the Relative displacement between the driven unit 701 and the mounting block 801 compensation mechanism shake is detected by using the detection position (see elements 422, 423 on the Hall effect in Fig. 2). Following this, in order to consider the relative displacement between the driven unit 701 and the fixing unit 801 as an observable output, the relative displacement between them is given as z0in the following formula (2):

[0066] state Variable is defined as follows:

[0067] Here, y=z0u=f, the absolute and the cosy speed fastener 801 is equal to w.

[0068] the Following formula (3) is obtained as the equation of state using the formula (1) and the above definitionname identities (where t is a variable that represents a time):

[0069] a Nu in the above formula represents the noise of the observations. This is a white Gaussian noise, and the mean value and the covariance of w and nu-known. w and nu should be represented by the following formula (4):

[0070] A-D and G in the formula (3) are represented by the following formula (5):

[0071] Therefore, taking into account the fact that the relative displacement between the driven unit 701 and the mounting block 801 compensation mechanism shake measurable, an observer is configured as shown in the following formula (6) in accordance with formula (3):

[0072] Here, L is the gain of the observer, which is the gain of the Kalman filter, which is previously obtained by solving the Riccati equation, shown in the following formula (7):

[0073] Here, L is determined by a positive definite symmetric solution P in the form of the following equation (8):

[0074] Using this monitor, be OC is the NCA absolute velocity induced block 701 mechanism for compensation of jitter (differential first order from z b), which is a state variable, and the relative displacement z0between excited by block 701 and the mounting block 801. By performing the differentiation of the first order on the estimated relative displacement z0and subtracting it from the absolute velocity induced block 701 to be detected absolute speed of the fixing unit 801, that is, the camera body (the differential of the first order from zw). In addition, if the differential of the first order estimated relative displacement z0is a very small value relative to the absolute velocity induced block 701, it can directly be used as the absolute velocity of the camera body to compensate for the translational shake. In this case, the above subtraction is not necessary, since dzbcan be considered equal to dzwbecause dz0it is small.

[0075] Fig. 6 is a block diagram showing an example of the configuration of the device to compensate for the jitter. In the following description, the configuration is shown only for jitter caused in the vertical direction (direction around the horizontal axis: see the arrows 103p, 104p in Fig. 1A) at the capture devices. Although a similar configuration is also provided for jitter caused in the horizontal direction (the direction of rotation in the Rog vertical axis: see arrows 103y, 104y in Fig. 1A) of capture devices, and describes only one of them, except for the difference in direction, essentially they are the same.

[0076] First described compensation for rotational shake.

[0077] the detection Signal of the angular velocity from the angular velocity meter is entered in the CPU 106 and processed in the integrated filter 901 HPF. The detection signal of the angular velocity is converted into an angular signal by discarding part of DC (direct current) using a HPF (high pass filter), which is integral filter 901 HPF, and then its integration. Band of the hand shake is from 1 to 10 Hz. Therefore, the HPF characteristics are characteristics that can cast component with a frequency sufficiently removed from the frequency band shake hands, for example, 0.1 Hz or less.

[0078] the Output of the integral filter 901 HPF is entered in block 903 sensitivity adjustment. Block 903 sensitivity adjustment increases the output of the integral filter 901 HPF to calculate the compensated target value of the rotational shake on the basis of information 902 position of the zoom and focus and the zoom photographing. This is intended to compensate for sensitivity compensation shake, changing the plane of formation of the image relative to the Ho is and the shake compensation unit 111 compensation shake, if the optical information is changed due to the adjustment of focus and zoom, etc. at the camera lens. Received compensated target value of the rotational shake is displayed in the exciting unit 110 through the block 912 addition, it uses the block 111 avoid the shakes and jitters can be compensated by the motion compensation lens. Device 905 assessment calculates the speed of parallel motion (speed translational motion) using output from the meter 108 angular velocity input with signal indications excitation from the excitation unit 110, and the output element on the Hall effect.

[0079] the following describes the compensation of translational shake.

[0080] As described above, in this embodiment, the speed of parallel motion (differential first order from zw) detected by the observer using a Kalman filter. As described in formulas (1) through (8), the absolute velocity induced block 701 and the relative offset between excited by block 701 and the mounting block 801 is estimated taking into account the relative displacement between the driven unit 701 and the mounting block 801 unit 111 of the shake compensation as a state variable and exciting efforts as an input variable.

[0081] Fig. 7 is a block diagram of a control, shows the surrounding configuration example device 905 valuation in Fig. 6.

[0082] Information about the relative offset of the derived elements on the Hall effect (see 422, 423 Fig. 2), and information about the force, indicated by signal indications excitation from the excitation unit 110 unit 111 compensate for jitter introduced into the filter 1001 Kalman. The signal specifying the excitation includes information related to the magnitude of the excitation unit 111 compensate for the jitter, and the exciting force is calculated in the following way. To convert the control current is specified by the controller 602 with feedback Fig. 4, in force, block 1002 transformation efforts calculates the exciting force by multiplying the current value by a predefined factor. The output of block 1002 transformation efforts is entered in block 1003 variable gain to compensate for temperature. The detection signal of the temperature (the temperature signal frame lenses) from a temperature sensor (not illustrated) provided next to the block 111 compensate for jitter, is entered in block 1003 variable gain. The rate of change of temperature, temperature-dependent structure of the lens, pre-stored in the storage device, and the gain (K) is variably controlled depending on the temperature change. Therefore, the speed of the parallel movement can be estimated taking into account temperature changes in the output characteristics of the coil BL is ka 111 compensate for the jitter due to temperature changes of the rim of the lens. Information about arousing effort, bred block 1003 variable gain, and information about the relative offset of the derived element on the Hall effect, are introduced into the filter 1001 Kalman. As described above, the filter 1001 Kalman estimates the absolute velocity induced block 701 of the block 111 compensate for jitter and relative offset between excited by block 701 and the mounting block 801. Differentiation of the first order is over valued relative offset in differentiating device 1004 to obtain the relative velocity, and is formed with an estimated absolute velocity induced block 701 in block 1005 addition. Through this calculates the estimated velocity of the translational shake (hereafter referred to as "estimated speed parallel motion" or "estimated speed of translational motion"). In addition, although the calculation performed in block 1005 addition, is actually a subtraction, addition (subtraction) for negative values should also be regarded as a similar calculation for convenience in the description, when it is not necessary to Express explicitly.

[0083] Returning to Fig. 6, describes the compensation of translational shake.

[0084] the Output of the meter 108 angular velocity is entered in the CPU 106 and processed in the integrated filter 909 HPF. After HPF, which is integral filter 909 HPF, threw DC component of the detection signal of the angular velocity, this signal is integrated and converted into an angular signal. The output of the integral filter 909 HPF is entered in block 910 gain control. Block 910 gain control contains the filter gain, and adjusts the amplitude and phase characteristics in the frequency band, which must be performed compensation of translational shake in conjunction with the process of integrated filter 909 HPF. The output of block 910 gain is compensated by the following block 911 compensation output and is seen as a compensated target value for the translational shake, is synthesized in block 912 addition and added to the compensated target value of the previously described rotational shake.

[0085] moreover, in parallel with the above process, the output of the meter 108 angular velocity is entered in block 904 adjustment phase HPF. HPF, which is a block 904 adjustment phase HPF, discards the DC component of the detection signal of the angular velocity, and then adjusts the phase of the signal. Only component with a frequency in a predetermined band is extracted from the output unit 904 adjustment phase HPF in block 906 BPF (band pass filter) angular velocity.

[0086] the Output (estimated near the banks of parallel motion device 905 assessment is sent in block 907 BPF speed parallel motion, and from it is extracted component with a frequency in a predetermined band. Each output unit 906 BPF angular velocity and unit 907 BPF speed parallel motion is entered in block 908 comparison, and calculates the compensation value (compensation coefficient), which compensates the output of block 910 gain control. Below will describe the calculation of the coefficient of compensation performed by block 908 comparison.

[0087] Information 902 position of the zoom and focus is also entered in block 911 compensation output and an increase in the photographing is calculated from this information. The output of block 910 gain is compensated on the basis of the obtained magnification photographing and amount of the compensation block 908 comparison, and calculates the compensated target value of the translational shake. Compensated target value of the translational jitter is added to the compensated target value of the rotational shake in block 912 addition. The result of the addition is introduced into the driving unit 110, and thereby excited the block 111 compensate for the jitter, and the jitter of the image will be compensated for the rotational shake and the translational shake.

[0088] the following describes the compensation value calculated by block 908 comparison.

[0089] Fig. 8 shows the rotational shake 103p and translational draw is the W 104p, attached to the capture device image, which is a schematic representation in side view. The amount of translational shake 104p in the position of the principal point of the optical system of the image-forming device 101 of the image capturing is recorded as Y. in Addition, the amount of rotational shake 103p written as theta, radius of rotation is written as L (see 1102p) when the center O of rotation (see 1101p). The ratio between them is represented by the following formulas (9) and (10):

[0090] Here, V is the velocity, and omega is the angular velocity. In addition, the radius L of rotation (see 1102p) is the distance from the center 1101p rotation to block detection of translational shake (in block 111 of the shake compensation).

[0091] In accordance with formula (9) integration of the first order is over estimated speed of parallel motion (estimated velocity of forward movement), the derived unit 905 ratings to get the Y offset, and the integration of the first order is executed on the output meter 108p angular velocity to get the angle theta, and the radius L of rotation is obtained from the value of their relationship. In addition, in accordance with formula (10) is estimated speed parallel movement, bred device 905 assessment, considered with the speed V, conclusion meter 108p angular velocity is regarded as the angular velocity omega and the radius L of rotation is obtained from the values of their ratio. The radius L of rotation can be obtained by another method.

[0092] When calculating the radius L of rotation peak value of the maximum amplitude of each of the velocity V and angular velocity omega within a predetermined time can be obtained to calculate the value of L from their ratio. A predetermined time is, for example, about 200 MS, if the cutoff frequency block 906 BPF angular velocity and unit 907 BPF speed parallel movement equal to 5 Hz. Moreover, the update of the radius L of the rotation can be performed all the time, when the calculated speed V and the angular velocity omega, respectively. In this case, component high-frequency noise can be removed when calculating the radius of rotation by performing coherent averaging over the velocity V and angular velocity omega, respectively, or drop high frequency component using a LPF (low pass filter).

[0093] From the offset Y translational shake in the position of the principal point of the optical system forming the image, the angle jitter theta, the focal length f of the optical system forming the image and zoom photographing beta increment the value of the jitter caused by the OE in the plane of image formation, is obtained by the following formula (11):

[0094] f and beta in the first member of the right-hand side of formula (11) obtained from information 902 position, zoom and focus, and the zoom photographing beta and the focal length f; and the angle jitter theta is obtained from the result of integrating the output of the meter 108p angular velocity. Therefore, the compensation for the rotational shake can be performed as described in Fig. 6. In addition, the second term in the right-hand side of formula (11) is obtained by performing integration of the first order on the estimated speed of parallel motion, derived device 905 assessment, and increase photographing beta derived from information 902 position, zoom and focus. Therefore, the compensation of translational shake can be performed as described in Fig. 6.

[0095] However, in this embodiment, the compensation image vibration is performed on the increment of value of jitter, where the formula (11) is rewritten as the following formula (12):

[0096] That is, the Y offset of the translational shake directly obtained by using the estimated speed of parallel motion, derived device 905 assessment is not used for the translational shake. Calculate the radius L of rotation obtained and the formula (9) or (10), and the offset value is calculated by multiplying this value of L, the result of integrating the output of the meter 108p angular velocity (theta) and increase photographing beta. The compensation value (beta, L), which is obtained by block 908 comparison in Fig. 6 using the calculated ratio of compensation and displayed in block 911 compensation output is the compensation coefficient with respect to theta.

[0097] Next, the management compensation of jitter is described with reference to Fig. 9A and 9B. The block diagram of the sequence of operations shown in Fig. 9A, begins with operation turn on the main power source device, image capture and executed in accordance with a program interpreted and executed by the CPU 106 in each fixed sampling period.

[0098] First, in step S1201 is determined to see if the user switch compensation shake (SW, not illustrated). The process proceeds to step S1202, if SW compensate for the jitter is in the enabled state, and proceeds to step S1219, if it is in the off state. At step S1202, the CPU 106 receives the detection signal from the meter 108 angular velocity. In the next step S1203, the CPU 106 determines whether it is in the state in which it is possible to compensate jitter; the process continues to E. the UPA S1204, if it is in the state in which the shake compensation is possible, and proceeds to step S1219, if he is not in a state in which it is possible to compensate jitter. At step S1203 is determined that he is not in a state in which it is possible to compensate jitter, since the power supply from the power source before you become stable the output meter 108 angular velocity. Is determined that it is in the state in which the shake compensation is possible, after the conclusion of the meter 108 angular rate became stable. Through this reduction of the efficiency of compensation of jitter can be prevented in a state in which the output immediately after the power source is unstable.

[0099] At step S1204 angle is calculated using the method described using Fig. 6. This angle is the output value of the integral filter 901 HPF in Fig. 6. Next, at step S1205 device 905 assessment calculates the estimated speed of the parallel motion with the aid of the method described above. Next, block 908 comparison calculates the radius of rotation for photographing a moving image at step S1206, and calculates the radius of rotation for photographing a still image at step S1207. At this point, the reasons for which are calculated separately rotational radii used the s when photographing a still image and the photographing of the moving image, are the following. Firstly, performing compensation of translational shake when photographing a moving image, the compensation lens unprofitable reaches the border of the range of movement (border control), if the impact of jitter applied to the plane of image formation of the translational shake, very large, for example, in the macroregion, etc. the Reason is that since the compensation range is limited, the compensation lens unprofitable tends to border control if the magnitude of the jitter is large. Therefore, the management of translational jitter is set to a lower level when photographing a moving image. For this reason, imposes a limit on the radius of rotation for photographing a moving image by changing the maximum value of the radius of rotation, etc., so that it was smaller than the radius of rotation for photographing a still image.

[0100] Next, in step S1208, the amount of compensation for the rotational shake is calculated on the basis of the information about the angle obtained in step S1204, information 902 position of the zoom and focus and the zoom photographing. Then, in step S1209, the CPU 106 determines whether the current mode is mode of photographing a moving image mode or photographer is of a still image; the process proceeds to step S1210, if this mode of photographing a moving image, and the process proceeds to step S1211, if this mode of photographing a still image. At step S1210, the compensation amount of translational shake for photographing a moving image is calculated using the radius of rotation for photographing moving images. Step S1211 is the process of determining the operational state of the button 105 descent (see Fig. 1). Button 105 descent supplied two-phase switches; poloniato button 105 descent puts the first switch (hereafter referred to as "SW1") in an enabled state, and a full button 105 descent causes the second switch (hereafter referred to as "SW2") in an enabled state. The CPU 106 checks whether the SW2 in the on or off state, the process proceeds to step S1212, if SW2 is in an on state, and the process proceeds to step S1213, if SW2 is in the off state.

[0101] In step S1212, the compensation amount of translational shake for photographing a still image is calculated using the radius of rotation for photographing a still image. In step S1213, the CPU 106 checks the status of SW1, the process proceeds to step S1214, if SW1 is in an on state is, and the process goes to step S1216, if SW1 is in the off state. In step S1214, the CPU 106 determines complete operation AF (auto focus). The process proceeds to step S1215, if the AF operation is completed, and the process proceeds to step S1216, if the AF operation is not completed.

[0102] At step S1215, the compensation amount of translational shake for photographing a moving image is calculated using the radius of rotation for photographing moving images. In addition, at step S1216, the compensation amount of translational shake is set to zero. After steps S1210, S1212, S1215, S1216 at step S1217 block 912, the addition (see Fig. 6) adds the compensation value of the rotational shake and the compensation value of the translational shake. At step S1218 exciting unit 110 outputs the control signal in block 111 of the shake compensation based on the calculated value of the shake compensation, and the compensation lens is excited. Meanwhile, at step S1219 stops the excitation of the compensation lens. After stages S1218, S1219 routine to compensate for jitter fails, and the standby process works until the next time discretization.

[0103] As described above, in the first embodiment, the absolute velocity induced block and the relative displacement between the driven unit and to epilim unit is estimated using an observer, where the relative displacement between the driven unit and the fixing unit compensation mechanism shake serves as a state variable, and stimulating effort to block compensation shake serves as the input variable. Then the estimated speed of the parallel movement is calculated to obtain the value of the compensation of translational shake. As described above, there is no need to provide accelerometer, etc. because it can be used to compensate for jitter and its a stimulant. Therefore, it is possible to realize downsizing and cost reduction, and compensation for the rotational shake and the compensation of translational shake can be performed simultaneously, without adding the component.

[0104] In the conventional configuration in which the acceleration sensor is used to detect the translational shake, you want the device to the attachment position of the acceleration sensor. Initially, the acceleration to be detected to compensate for the translational shake, is the acceleration at the position of the principal point of the lens. However, the acceleration sensor must be attached to a location remote from the position of the principal point of the lens and the center of the optical axis (substrate control in the camera body, and so on), if there is not sufficient space around the lens and attaching the acceleration sensor is difficult. For this reason, there is a fear that the detection accuracy of the detection signal of the acceleration and the magnitude of the translational shake will be reduced. In contrast, in this embodiment, since the amount of translational shake can be calculated for a portion of the mechanism for compensation of jitter, the amount of translational shake is calculated about the position of the principal point of the lens and the center of the optical axis, and the accuracy of the compensation of translational jitter may be increased.

[0105] in Addition, in this embodiment, the described movement of the compensation lens as a means of compensation of jitter in the plane perpendicular to the optical axis, that is what is called "optical shake compensation". However, it can be used not only for optical compensation of jitter, but also the following configuration:

the configuration in which the shake compensation is performed by moving the imager in the plane perpendicular to the optical axis;

configuration using electronic compensation of jitter, which reduces the effects of jitter by changing the position of the clipping image of each photographic frame, the extracted element in the formation of the image; and

the configuration in which the shake compensation is performed by merge is of many management tools compensation shake.

[0106] in Addition, each type of configuration it is also possible to estimate the translational shake. For example, in the case of using the mechanism of compensation of jitter, which compensates the jitter by moving the imager in the plane perpendicular to the optical axis, the estimated speed of the parallel movement can be calculated by the observer, if it can be done monitoring the relative displacement between the driven unit and the fixing unit compensation mechanism shake. That is the absolute velocity of the camera body is detected by using a method similar to the above, if the observer is configured so that the relative displacement between the driven unit and the fixing unit compensation mechanism shake serves as a state variable, and stimulating effort to block compensation shake serves as the input variable.

[0107] [Second variant implementation]

[0108] the following describes a second variant implementation of the present invention.

[0109] Fig. 10 shows an example of a device configuration of the shake compensation in accordance with the second embodiment. In addition, skipping the detailed description of components that are the same as in the first embodiment, using ispolzovanie characters and mainly describes the difference between the first embodiment and the second embodiment. The same applies to other variants of implementation, are described below.

[0110] In the second embodiment, compensation of translational shake is performed based on the signal of the parallel displacement obtained by integrating the estimated speed parallel motion calculated by the device 905 assessment. In addition, the on/off compensation of translational shake is set depending on the state of the second switch SW2, and the detection signal meter 108 angular velocity is determined whether a predetermined or greater change in postural position just before the operation SW2, and the on/off compensation of translational shake installed during the photographing operation.

[0111] Therefore, the differences between Fig. 6 and 10 are the following:

(1) refer to operational information 1301 SW2;

(2) there is no integrated filter 909 HPF, block 910 gain control block 911 compensation output unit 904 adjustment phase HPF, block 906 BPF angular velocity, unit 907 BPF speed parallel motion, block 908 comparison, shown in Fig. 6; and the integral filter 1302 HPF and block 1303 sensitivity adjustment feature for Vici the population size of compensation of translational shake;

(3) is the unit 1304 switching signal according to the operational information 1301 SW2; and

(4) feature block 1305 determine changes in position and block 1306 switching signal at its output.

[0112] As compensation for the rotational shake is similar to those in the first embodiment, the following describes the compensation of translational shake.

[0113] the Estimated speed of the parallel movement of the device 905 assessment is introduced in the integral filter 1302 HPF. HPF, which is integral filter 1302 HPF, discards the DC component of the input signal, and this signal is converted into a parallel signal offset by integration and is entered in block 1303 sensitivity adjustment. Unit 1303 sensitivity adjustment increases the output of the integral filter 1302 HPF to calculate the compensated target value of the rotational jitter based on the magnification of the photographing derived from information 902 position, zoom and focus.

[0114] the Operational information 1301 SW2 is sent in block 1304, the switching signal. Block 1304 switching signal selects zero or the output unit 1303 sensitivity adjustment and outputs the result to a subsequent block 1306 switching signal, with reference to the operational information 1301. That is, if SW2 is in an on state, then you selected the d block 1303 sensitivity adjustment, and block 1304 switching signal displays the amount of translational shake. Besides, if SW2 is in the off state, the unit 1304 switching signal selects zero, and no compensation of translational shake is not performed. The reason is that since the evaluated device 905 evaluation speed parallel movements may significantly affect the acceleration of free fall caused due to the change of the position of the capture devices, erroneous estimation of the speed of parallel motion adversely arise due to the influence of the acceleration of gravity, if the position change is significant. In many cases, because the photographer does a great change in the operation SW2, a suitable calculation of the translational shake possible on SW2, and performs compensation of translational shake.

[0115] Although the unit 1305 determine changes position determines the position change device capture images using angular signal, which is output to the integral filter 901 HPF, it calculates the change in angular signal at a predetermined time during each control sample. Besides operational information 1301 SW2 is also entered in block 1305 determine position changes; while SW2 is in the on mode the when condition, information defining the position held immediately before SW2 is switched from an off to an on state. This is intended to prevent a decrease in the efficiency of the management shake compensation using the on/off state of the switching compensation of translational shake until SW2 is in an on state, that is, in the photographing operation.

[0116] Block 1306 switching signal selects and outputs zero or the output unit 1304 switching signal depending on the output unit 1305 determine position changes. That is, if the change in position is determined by the block 1305 determine changes position as greater than or equal to the threshold value, when SW2 is pressed, the unit 1306 switching signal selects zero in accordance with the output that indicates the result of the determination. Moreover, if the position change is defined as a lower threshold, when SW2 is pressed, then selects the output unit 1304 switching signal.

[0117] the Operational information 1301 SW2 is also entered in the integral filter 1302 HPF, and the output of the integral filter 1302 HPF initialized to zero due to the enabled state of the SW2. This is intended to prevent disadvantageous step change of magnitude compensate for jitter introduced in the driving unit 10, immediately after performing compensation of translational shake; the compensation value of the translational shake is initialized, if SW2 is switched from an off to an on state.

[0118] Fig. 11 is a diagram illustrating the operation by changing the position and operation SW2 to compensate for the translational shake. Shows the temporal variation in the output block 1401 1303 adjust the sensitivity and temporal variation in the output signals 1402, 1403 (value compensation of translational shake) block 1306 switching signal. Signal 1402 shown in Fig. 11A indicates a change that is equal to zero, while SW2 is not pressed, and is similar to the conclusion 1401 after pressing. In addition, as shown in Fig. 11B signal 1403 indicates zero.

[0119] In the case of Fig. 11A is compensation of translational shake, because a large position change does not occur when SW2 is pressed. SW2 is in the off state in preparation for the photographing, and the level of signal 1402 indicates zero. The compensation value of the translational shake initialized to zero at the moment of pressing SW2, and calculates the offset shift with zero initial value. Signal 1402 indicates the level obtained by subtracting the portion offset from the output 1401, and the compensation value of the translational jitter is calculated by the subtraction of this is moving, until ends the photographing operation. In contrast, the compensation of translational shake is set so as not to run if there is a big change of position, when SW2 is pressed, as shown in Fig. 11B. The signal 1403 is set to zero, when SW2 is pressed, and the compensation value of the translational shake is maintained at zero until ends the photographing operation.

[0120] In the second embodiment, although the compensation of translational shake can be performed only when SW2 is pressed (in the operation of photographing a still image), it can be combined with the method of compensation of translational shake in the first embodiment. That is, the above-described compensation of translational shake occurs in the operation of photographing a still image by pressing SW2. In addition, the compensation of translational shake from the radius of rotation described in the first embodiment, is performed in preparation for photographing before SW2 is pressed, and when photographing a moving image. Therefore a suitable compensation of translational shake possible depending on the situation.

[0121] In accordance with a second embodiment of the installation on/off switch can be is switched to compensate for the translational shake depending on the operating state of the SW2, that is, on whether the button 105 descent into the fully pressed state. This enables compensation of translational shake along with the exception of the influence of the acceleration of free fall in the operation of photographing. Therefore, the efficiency of compensation of jitter in the operation of photographing a still image increases, because more accurate compensation of translational shake can be performed compared with the compensation of translational shake, using the radius of rotation described in the first embodiment.

[0122] [Third option exercise]

[0123] the following describes a third option of implementing the present invention.

[0124] Feature, which a compensation of jitter in accordance with the third embodiment differs from the device to compensate for jitter in accordance with the sixth embodiment of the present invention is the configuration of device evaluation. In the third embodiment, the output of the meter 108 angular velocity is introduced into the device 1501 ratings together with signal indications excitation from the excitation unit 110 and the output element on the Hall effect, as indicated by the dashed arrow in Fig. 6.

[0125] Fig. 12 is a block diagram showing the internal process in the device 1501 evaluation. Device for estimating the rate of couples who thelego movement in the third embodiment is different from the configuration example, it is shown in Fig. 7, in the following respects:

(1) is the unit 1601 determine the condition of oscillation, which determines the state of oscillation, so that the output meter 108 angular velocity is used as input; and

(2) is the block 1604 determine gain Kalman, in which the output unit 1601 determine the status of the oscillations is used as input, the output of which is sent to the filter 1001 Kalman, and the speed of the parallel movement is measured using the gain Kalman installed in block 1604 determine gain Kalman.

[0126] the Process of calculating the estimated speed of the parallel movement is described below.

[0127] Block 1601 determine the status of the fluctuations calculates the amount of state oscillations by obtaining the output of the meter 108 angular velocity, and outputs the result in block 1604 determine gain Kalman. In block 1601 determine the status of the oscillations of the high frequency component is discarded unit 1603 processing LPF after the conclusion of the meter 108 angular velocity converted to an absolute value using block 1602 processing of the absolute value. Processed by the LPF signal is calculated as a residual state fluctuations.

[0128] the Process block 1610 determine the status of the fluctuations describes what I'm referring to Fig. 13A and 13B. Fig. 13A illustrates the temporal variation in the output 1701 meter 108 angular velocity and the output block 1702 1602 processing of the absolute value. Fig. 13B illustrates the temporal variation in the output 1703 block 1603 LPF processing.

[0129] the Output 1701 meter 108 angular velocity shown in Fig. 13A is an absolute value, which is transformed in block 1602 processing of the absolute value, the result is a conclusion 1702, and further the high-frequency component is discarded in block 1603, the LPF processing. Cutoff frequency of the LPF, which is a block 1603 processing LPF is set, for example, at a frequency of 0.5 Hz or less, and the result is output 1703 shown in Fig. 13B. In addition, the block 1603 processing LPF may be a configuration in which the moving average is calculated in a predetermined period.

[0130] Among the periods TA-TD, as shown in Fig. 13B, the period TB shows the state in which the hand shake is very strong. In this case, the output unit 1603 LPF processing is displayed so that it remains a great value. In addition, the period TD shows the state in which the hand shake is weak. In this case, the output unit 1603 LPF processing is displayed so that it remained a small value.

[0131] the Output unit 1603 LPF processing, that is, the output unit 1601 determine the status of the oscillations, is entered in block 1604 determine the gain To the of Lana. Block 1604 determine the gain of the Kalman selects the gain of the Kalman installed in accordance with the state of oscillation, and sets the filter parameters 1001 Kalman. In block 1604 determine the gain of the Kalman threshold values determine the status of the fluctuations installed as Th3, Th2, Th1, as shown in Fig. 13B. By comparing the output value of the block 1603 LPF processing with these thresholds determine the gain of the Kalman determined depending on the comparison result, which specifies in what range it is in. That is, in this example, the values of gain Kalman stored in the storage device for ranges of more than Th3, between Th3 and Th2, between Th2 and Th1 and smaller than Th1, respectively.

[0132] To a suitable speed parallel motion can be assessed depending on the condition of oscillation, the gain of the Kalman depending on the size of the state fluctuations. By early prediction of interference depending on the size of the state fluctuations obtained from the angular speed can be pre-set gain Kalman. In this example, the status of the fluctuations is performed by using the angular velocity. For example, this definition States fluctuation etc which leads to a suitable value for the evaluation of translational shake, if the angular velocity is very large, and the speed of the parallel movement is very small, and if the angular velocity is very small, and the speed of the parallel movement is very large. However, in the case of a conventional hand-held photography sometimes a situation in which the rotational shake is very weak, and the translational shake is very strong, and the amount of state oscillations obtained by the angular velocity associated with interference. Therefore, by changing the gain Kalman, who has previously installed, depending on the size of the state fluctuations obtained from the angular velocity suitable estimated speed of parallel motion is obtained depending on the situation fluctuations, and increases the efficiency of the shake compensation using compensation of translational shake.

[0133] [Fourth option exercise]

[0134] the following describes a fourth variant of implementation of the present invention.

[0135] Fig. 14 is a block diagram showing a configuration example of device 1501 evaluation device for compensation of jitter in accordance with the fourth embodiment. Differences between Fig. 12 and 14 are the following:

(1) instead of block 1604 determine gain Kalman is a block 1801 determine the input gain for the control blocks 1802, 1803 input variable is the first gain; and

(2) blocks 1802 and 1803 input variable gain amplify the output (relative offset) element on the Hall effect and the signal indications excitation, respectively.

[0136] In the fourth embodiment, the output unit 1601 determine the status of the oscillations is entered in block 1801 determine the input gain. Block 1802 input variable gain amplifies the output (relative offset) element on the Hall effect and outputs the result to the filter 1001 Kalman in accordance with the output of block 1801 determine the input gain. In addition, the unit 1803 input variable gain amplifies the signal directions of excitation and displays the result in block 1002 transformation efforts in accordance with the output of block 1801 determine the input gain. Therefore the estimated speed of the parallel movement, the calculated filter 1001 Kalman, is large, if the output unit 1601 determine the status of the oscillations is large, and the estimated speed of the parallel movement, the calculated filter 1001 Kalman is low, if the output unit 1601 determine the status of the oscillations is small.

[0137] To a suitable speed parallel motion can be assessed depending on the condition of oscillation, the gain of the input signal to the filter 1001 Kalman depending on the size of the state fluctuations, and block determine the input gain supports the amplification value. Although the estimated value of the filter 1001 Kalman can be used directly if the oscillation is very strong, the gain is set so that the estimated value of the filter 1001 Kalman was small, if the oscillation is very weak. This is why the concern about erroneous assessment occurs in which, when the oscillation is weak, the estimation accuracy of the estimated value of the velocity parallel movement of the Kalman filter is low. Namely, if the estimated speed of the parallel movement is set to a large value, even if the actual speed of the parallel movement is small, the efficiency of compensation shake disadvantageously reduced due to excessive compensation at compensation of translational shake. To prevent this, is the above-described setting of the gain. In the case of a conventional hand-held photography, because there are some situations in which the rotational shake is very weak, and the translational shake is very strong, the compensation amount of translational shake is limited so that it is not too big, if the rotational shake is very weak.

[0138] In accordance with the fourth embodiment, by changing the gain of the input signal to the filter 1001 Kalman, which has been previously installed, depending on the size of the state oscillations obtained from a signal about what Eugenia angular velocity, can be prevented by reducing the effectiveness of the shake compensation at compensation of translational shake due to erroneous estimation of the speed of parallel motion.

[0139] the [Fifth variant implementation]

[0140] the following describes the fifth implementation of the present invention.

[0141] Fig. 15 is a block diagram showing a configuration example of device 1501 evaluation of devices compensating jitter in accordance with the fifth embodiment. Differences between Fig. 12 and 15 are the following:

(1) instead of block 1604 determine gain Kalman is a block 1901 installation boundary frequency tunable filter to control HPF tunable 1902 and 1903; and

(2) HPF tunable 1902 and 1903 output processed by the HPF signals in output (relative offset) element on the Hall effect and the signal indications excitation, respectively.

[0142] In the fifth embodiment, the output unit 1601 determine the status of the oscillations is entered in block 1901 installation boundary frequency tunable filter (hereinafter called the "unit setup cutoff frequency"), the output of which is sent to HPF tunable 1902 and HPF tunable 1903 respectively. Tunable HPF 1902, which processes the output (relative offset) element on the Hall effect, is Graniczna the frequency, changed depending on the output unit 1901 installation cutoff frequency, and outputs the filtered signal to the filter 1001 Kalman. Besides HPF tunable 1903, which processes the signal specifying the excitation has a cutoff frequency that is changed depending on the output unit 1901 installation cutoff frequency, and outputs the filtered signal in block 1002 transformation efforts. In accordance with the installation boundary frequency depending on the magnitude of the state fluctuations in block 1901 installation boundary frequency estimated speed parallel motion calculated by the filter 1001 Kalman, calculated without low-frequency amplification, cast, if the output unit 1601 determine the status of the oscillations is large. Moreover, the estimated speed of the parallel movement is calculated with the low-frequency gain, reducing, if the output unit 1601 determine the status of the oscillations is small.

[0143] To a suitable speed parallel motion can be assessed depending on the condition of oscillation, the cutoff frequency of each HPF tunable output (relative offset) element on the Hall effect and the signal specifying the excitation installed depending on the size of the state fluctuations. Through this assessment appropriate speed parallel motion nightclub is on depending on the size of the state fluctuations. If the value of the state fluctuations greater than or equal to the threshold value, the cutoff frequency of the HPF tunable is set to a lower level, and can be used directly assessed filter 1001 Kalman value. However, if the value of the state fluctuations less than the threshold value, then the low-frequency gain in appreciated value of the filter 1001 Kalman designed small by setting the cutoff frequency HPF tunable in a large. This is why the concern about erroneous evaluation occurs when the oscillation is weak in the case in which the estimation accuracy of the velocity parallel motion filter 1001 Kalman low. Namely, if the estimated speed of the parallel movement is calculated in a large value even if the actual speed of the parallel movement is small, the efficiency of compensation shake disadvantageously reduced due to excessive compensation at compensation of translational shake. Therefore, excessive compensation can be prevented by using the above setup, the cutoff frequency HPF tunable. In the case of a conventional hand-held photography, because there are some situations in which the rotational shake is very weak, and the translational shake is very strong, the filter configuration is performed in order to prevent an error in the calculation of the low-frequency characteristics of the assessed value of compensation of translational shake, if the rotational shake is very weak.

[0144] In accordance with the fifth embodiment by changing the cutoff frequency HPF tunable, which processes the input signal to the filter 1001 Kalman, depending on the size of the state fluctuations obtained from the angular velocity, it is possible to prevent erroneous estimation of the speed of parallel motion, and to increase the effectiveness of the shake compensation at compensation of translational shake.

[0145] [Sixth variant implementation]

[0146] the following describes a sixth variant of implementation of the present invention.

[0147] Fig. 16 shows an example of a device configuration of the shake compensation in accordance with the sixth embodiment.

[0148] In the sixth embodiment, compensation for the rotational shake and the compensation of translational shake is not performed in arousing unit 110 unit 111 compensate for jitter, as in the first embodiment, and compensation for the rotational shake and the compensation of translational shake performed separately. That is, compensation for the rotational shake is stimulating unit 110 unit 111 of the shake compensation, and compensation of translational shake is a block 2001 compensation shake with clipping images. Unit 2001 compensation shake with clipping image is of performs the compensation process shake by shifting the output area in the photographic image, formed from the output of element 107 of image formation, in accordance with the compensation value of the translational shake, indicated by block 911 compensation output.

[0149] the Difference between Fig. 6 and 16 is the following:

(1) in Fig. 16 no block 912 addition, as shown in Fig. 6, and the output unit 903 sensitivity adjustment is entered in the exciting unit 110; and

(2) the output of the block 911 compensation output is entered in block 2001 compensation shake with clipping of the image.

[0150] Fig. 17 is a diagram illustrating how the process of shake compensation is performed by shifting the output area in the photographic image formed with the output element 107 imaging, using the process in block 2001 compensation shake with clipping of the image.

[0151] Fig. 17 shows the output image 2101a of element 107 form the image at the time t1 and the output image 2101b of element 107 form the image at the time t2 after it has passed the predetermined time (for example, 1/30 seconds after). If not provided the remedy jitter, which compensates for the shake of the image by moving the photographic optical axis, the two images differ from each other in composition due to the rotational shake and the translational shake. At stake is Horatii in Fig. 16, as compensation for the rotational shake is performed in arousing unit 110 unit 111 compensate for the jitter, and the rotational shake is compensated, the composition will differ from each other due to the remaining translational shake.

[0152] the Block 911 compensation output displays the compensation value of the translational shake in each of the horizontal and vertical directions in the unit 2001 compensation shake with clipping images. Unit 2001 compensation shake with clipping image calculates the amount of movement (see arrow 2102y, 2102p) trim the image in each of the horizontal and vertical directions for each photographic frame, and moves (shifts) the position of the cut-off images on the amount of movement (see arrow 2102) trim the image. That is, the arrow 2102 represent the amount of movement corresponding to the translational shake and the vector in the direction of compensation. This causes the range of the clipping image at the time t1 photographing image 2101a to the range shown in image 2013. This also causes the range of the cut image in the image 2101b at t2 photography to the range shown in image 2013, and the photographing of the moving image can be performed on the flower (see image 2104), which is what I main photographic object without image vibration. As a result of performing compensation of translational shake clipping path image for each photographic frame compensation of translational shake can be performed concurrently with compensation for the rotational shake when photographing a moving image.

[0153] In the sixth embodiment, to compensate for the rotational shake tool is used to compensate for jitter, which compensates for the shake of the image by moving the photographic optical axis. In addition, to compensate for the translational shake tool is used to compensate for jitter by clipping the image, which performs the compensation process jitter by changing the output area of the photographic image. Through this range of excitation to compensate for the rotational shake and the range of excitation to compensate for the translational shake can be provided accordingly. Therefore, because the range of excitation to compensate for image vibration can be increased, an unexpected decrease in the efficiency of management compensation shake can be prevented in the vicinity of border control due to failure of the control range compensation shake.

[0154] [Seventh variant implementation]

[0155] Next, described is ivalsa seventh variant of implementation of the present invention.

[0156] Fig. 18 is a block diagram of the control showing an example of the configuration of the device 905 assessment. The device estimates the speed of parallel motion in the seventh embodiment is different from the configuration example shown in Fig. 7, in the following respects:

(1) is the unit 2206 definition pan and regulations, which specifies the pan and change of position, so that the output meter 108 angular velocity is used as input; and

(2) is the unit 2207 determine gain Kalman, in which the output unit 2206 definition pan and position serves as the input, the output of which is sent to the filter 1001 Kalman, and the speed of the parallel movement is measured using the gain Kalman installed in the unit 2207 determine gain Kalman.

[0157] the Process of calculating the estimated speed of the parallel movement is described below.

[0158] the Output of the meter 108 angular velocity is entered in block 2206 definition pan and regulations, and the definition of the pan and change the position, i.e. the result of the comparison of the magnitude of the position change and the threshold values entered in block 2207 determine gain Kalman. Through this, the gain of the Kalman ustanavlivat the Xia depending on the result of the determination of block 2206 definition pan and regulations and changed the estimated speed of the parallel motion.

[0159] the Process up until not rated speed parallel movement on the basis of the definition pan in block 2206 definition pan and conditions described below using Fig. 19A and 19B.

[0160] Fig. 19A shows the temporal variation in the output 2301 meter 108 angular velocity. Fig. 19B shows the temporal variation in estimated speeds 2302, 2303 parallel motion. Estimated speed 2302 parallel motion is the speed of the parallel motion, the estimated filter 1001 Kalman, if the gain of the Kalman change unit 2207 determine gain Kalman. Estimated speed 2303 parallel motion is the speed of parallel motion, estimated from filter 1001 Kalman, if the gain of the Kalman does not change the block 2207 determine gain Kalman. The periods TA and TD in Fig. 8 is a period during which the user is in a normal state hand-held shooting without running panning. In this case, since the change in position is less than the threshold value, the block 2206 definition pan and position outputs the signal definition, which specifies the panning does not work". In addition, the period TB is a period of the on time of the operation of the pan, and block 2206 definition pan and position displays definition signal, which indicates "operation pan". The period TC is the period directly before and directly after the operation of the pan and the block 2206 definition pan and position displays definition signal, which indicates the completion of the pan".

[0161] the Block 2206 definition pan and regulations determines that the device performs the operation of the pan and is subject to change in position, if the detected angular velocity continues to be greater than or equal to the threshold value within a predetermined time or longer. Changing the position before and after the moment when the pan operation is completed, will also be included in the last position change. Either unit can determine that the device performs the operation of the pan and is subject to change in position, if the angular acceleration, which is a first order differential of the detected angular velocity, continues to be greater than or equal to the threshold value within a predetermined time or longer. Either unit can determine that the device performs the operation of the pan and is subject to change positions, based on what is detected Uglova the rate continues to be greater than or equal to the threshold value within a predetermined time or longer, and angular acceleration continues to be greater than or equal to the threshold value within a predetermined time or longer.

[0162] In the periods TB and TC gain Kalman obtained earlier, so that the estimated speed of the parallel motion, which is the result of the calculation is small. In particular, it is necessary to prevent erroneous estimation of velocity parallel movements in the periods TC and TD. Gain Kalman, which corresponds to the usual hand shake, is installed outside of the pan in the period TA and TD.

[0163] the Control by which the estimated speed of the parallel movement is intentionally made smaller, is accomplished by changing the gain under the condition in which the speed of the parallel movement becomes much more as a result of change of acceleration of falling in connection with the repositioning device and the external accelerations during the operation of the pan. Through this it is possible to prevent excessive compensation compensation of jitter in accordance with the compensation of translational shake during and immediately before and immediately after the operation of the pan and change the position.

[0164] Hereinafter explained all the work on management compensation of jitter in accordance with the seventh var is the ant, with reference to Fig. 20A and 20B. A flowchart of sequences of operations shown in Fig. 20A and 20B, starting from the operation to turn on the main power source device, image capture, and are each fixed sampling period in accordance with the program that is interpreted and executed, the CPU 106.

[0165] the Seventh variant implementation differs from the first variant implementation in the following respect.

[0166] in Step S2405, the following S1204, is the process of determining the panning performed by block 2206 definition pan and position, and proceeds to step S1307, if it is determined that it does not pan operation, and proceeds to step S2406, if it is determined that it performs an operation of the pan. In step S2406, the device settings 905 valuation changes, as described in Fig. 19A and 19B, and the process proceeds to step S1205. Above the previous and subsequent processes (S1201-S1204 and S1205-S1219) are similar to those in the first embodiment.

[0167] As described above, in the seventh embodiment, can be calculated speed parallel motion, and can be performed compensation of translational shake. It is assumed that the photographer performs photographing operation depending on the time along with cropping by following photographic object, or that the pictures the f performs an operation of photographing along with adjustment of the jitter angle caused by hand shake, etc., In this case there is also the influence of translational shake in relation to shake hands, which is not planned by the photographer, the influence of the translational shake is great due to the fact that the photographer intentionally shaking the camera. If the compensation of translational shake from the radius of rotation in accordance with the compensation for the rotational shake, the radius of rotation during the operation of the pan is a great value, and you might make a mistake in the estimation of the radius of rotation when photographing immediately after the operation of the pan. That is, if the compensation value of the translational shake became excessive when photographing immediately after the operation of the pan and change the position, it is disadvantageous effect on the result of the shake compensation at compensation shake. Therefore, in this embodiment, by changing the gain Kalman, which is previously set depending on the result of determination of pan and regulations, you can prevent overestimation of the speed of parallel motion during the operation of the pan and changing the position. In addition, in the operation of the pan translational jitter becomes large due to the fact that the photographer intentionally shaking the device capturing the changes the position of the device in a certain direction. In this case, since there is no need to perform compensation of translational shake, managed, by which the estimated speed of the parallel movement is intentionally made smaller, so that the compensation amount of translational shake in the operation of the pan was small.

[0168] according To the seventh embodiment of the effectiveness of the shake compensation at compensation translational jitter is improved by preventing a decrease in the efficiency of compensation of jitter due to excess compensation of translational shake during and immediately after the pan and change the position.

[0169] [Eighth variant implementation]

[0170] the following describes the eighth variant of implementation of the present invention.

[0171] Fig. 21 is a block diagram showing an example of the configuration of the device 905 evaluation device for compensation of jitter in accordance with the eighth embodiment. The differences between the configurations shown in Fig. 18 and 21, are as follows:

(1) instead of block 2207 determine the gain of the Kalman feature block 2501 determine the input gain and blocks 2502 and 2503 input variable gain; and

(2) blocks 2502 and 2503 input variable gain amplify the output (relative offset) element on the Hall effect and the persecuted instructions excitation, respectively.

[0172] In the eighth embodiment, the output unit 2206 definition pan and position is entered in block 2501 determine the input gain. The output unit 2501 determine the input gain is sent to each of the blocks 2502, 2503 input variable gain. Block 2502 input variable gain amplifies the output (relative offset) element on the Hall effect and outputs the result to the filter 1001 Kalman in accordance with the output of block 2501 determine the input gain. In addition, the block 2503 input variable gain amplifies the signal directions of excitation and displays the result in block 1002 transformation efforts in accordance with the output of block 2501 determine the input gain.

[0173] Through this estimated speed parallel movement in the filter 1001 Kalman is small, if the block 2206 definition pan and conditions is determined that the device performs the operation of the pan and is subject to change positions. Namely, if there is a state of the pan and changing the position of the device, the amplification unit 2502 or 2503 input AC gain is set by the output unit 2501 determine the input gain so that the estimated speed of the parallel movement was small in comparison with a case where it is determined that the device does not perform operas is on the pan. In addition, if the block 2206 definition pan and conditions is determined that the device is not performing a pan operation and is subject to change in position, the unit 2501 determine the input gain controls unit 2502 or 2503 input variable gain, and the gain in accordance with the magnitude of the normal hand shake.

[0174] Each of the gain blocks 2502 and 2503 input variable gain is set for the intentional decrease estimated speed parallel motion under the condition in which the speed of the parallel movement becomes very large due to the influence of the acceleration of gravity in connection with the repositioning device and the influence of the perturbing acceleration due to the operation of the pan.

[0175] In accordance with the eighth embodiment of the effectiveness of the shake compensation at compensation translational jitter is improved by preventing a decrease in the efficiency of compensation of jitter due to excessive compensation at compensation of translational shake during and immediately after the pan and change the position.

[0176] [Ninth variant implementation]

[0177] the following describes the ninth variant of implementation of the present invention.

[0178] Fig. 22 is a block diagram showing an example of the configuration of the device 905 assessment in the equip compensation of jitter in accordance with the ninth embodiment. The differences between the configurations shown in Fig. 18 and 22, are the following:

(1) instead of block 2207 determine the gain of the Kalman feature block 2601 installation boundary frequency tunable filter and HPF tunable 2602 and 2603; and

(2) HPF tunable 2602 and 2603 output processed by the HPF signal in each of output (relative offset) element on the Hall effect and signal directions of excitation.

[0179] In the ninth embodiment, the output unit 2206 definition pan and position is entered in block 2601 installation boundary frequency tunable filter (hereinafter called the "unit setup cutoff frequency"). The output unit 2601 installation cutoff frequency is entered in HPF tunable 2602 and 2603, and the cutoff frequency of the HPF process is changed with respect to each of output (relative offset) element on the Hall effect and signal directions of excitation. Through this cutoff frequency of the HPF is set at a high frequency, if the block 2206 definition pan and regulations determines that the device performs the operation of the pan and is subject to change positions. That is, the gain that is less than or equal to the boundary frequency of the HPF is set at a lower level for the estimated speed of parallel motion calculated by the filter 1001 Ka is mana. In addition, if the block 2206 definition pan and conditions is determined that the device is not performing a pan operation and is subject to change in position, the cutoff frequency of the HPF is set at a lower level and is set to enhance performance in accordance with the magnitude of the normal hand shake.

[0180] the cutoff frequency HPF tunable 2602 and 2603 are intended to reduce the estimated speed parallel motion under the condition in which the speed of the parallel movement becomes very large due to the influence of the acceleration of gravity in connection with the repositioning device and the influence of the perturbing acceleration due to the operation of the pan.

[0181] In accordance with the ninth embodiment of the effectiveness of the shake compensation at compensation translational jitter is improved by preventing a decrease in the efficiency of compensation of jitter due to excessive compensation at compensation of translational shake during and immediately after the pan and change the position.

[0182] [Tenth variant implementation]

[0183] the following describes the tenth variant of implementation of the present invention.

[0184] Fig. 23 is a block diagram showing an example of the configuration of the device 905 evaluation device for compensation of jitter in accordance with ezetim of the embodiment. The difference between the configurations shown in Fig. 22 and 23, is the following:

(1) block 2701 calculation, which calculates the estimated speed of the parallel movement in the control period (1 sampling period) by 1 before, and block 2702 definition pan and conditions are provided instead of the meter 108 angular velocity and block 2206 definition pan and regulations; and

(2) the estimated speed of the parallel movement in the period of control by one unit before you enter from the block 2701 calculations in block 2702 definition pan and conditions.

[0185] the Tenth version of the implementation differs from the ninth variant implementation in the following respect.

[0186] the Estimated speed of the parallel motion, which is the output of block 1005 addition, is entered in block 2701 calculations, which stores the estimated speed of the parallel motion to calculate the next control period. That is, the block 2701 computing is a means of delay, which holds an estimated speed of parallel motion in the previous control period, the output of which is sent in block 2702 definition pan and regulations, and determines whether the device is a pan operation and whether the change of position. The result of the determination is entered in block 2601 installation Gran who offered frequency. The subsequent processes are similar to those in the ninth embodiment.

[0187] the Process up until the estimated speed of the parallel movement of the definitions of the pan in block 2702 definition pan and conditions described below using Fig. 24.

[0188] the Estimated speed 2801 parallel movements shown in Fig. 24, is the speed of the parallel motion, the estimated filter 1001 Kalman, if the block 2601 installation cutoff frequency does not change the cutoff frequency HPF tunable 2602 and 2603. Moreover, the estimated speed 2802 parallel motion is the speed of the parallel motion, the estimated filter 1001 Kalman, if the block 2601 installation cutoff frequency does not change the cutoff frequency HPF tunable 2602 and 2603 in response to the result of the determination from block 2702 definition pan and conditions.

[0189] In the period TA in Fig. 24 user is in a normal state hand-held shooting without performing pan, block 2702 definition pan and position outputs the signal definition, which specifies the panning does not work". In the period TB is determined that the device performs the operation of the pan and is subject to change positions based on the estimated speed of parallel motion. Block 2702 definition pan and p is the situation determines the device performs the operation of the pan and is subject to change in position, if the estimated speed of the parallel movement continues to be greater than or equal to the threshold value within a predetermined time or longer. Either unit may also determine that a device is performing a pan operation and is subject to change if the estimated acceleration of parallel motion, which is the differential of the first order estimated speed parallel motion, continues to be greater than or equal to the threshold value within a predetermined time or longer. Alternatively, the unit can determine that the device is in operation the pan and changing the position, if the estimated speed of the parallel movement continues to be greater than or equal to the threshold value within a predetermined time or longer, and the estimated acceleration parallel movement continues to be greater than or equal to the threshold value within a predetermined time or longer.

[0190] the cutoff frequency of the HPF tunable 2602 and 2603 are set at a higher level than usual, block 2601 installation cutoff frequency in a certain period, if it is determined that the device is performing a pan operation and undergoes change the situation. In the example of Fig. 14 speed parallel motion is estimated in accordance with the magnitude of the normal hand shake while not working pan shown in the periods TA and TC. In the operation of the pan shown in the period TB, the cutoff frequency of the tunable filter HPF is set at a higher level than usual, and the estimated speed of the parallel movement is controlled to be small. For this purpose it is necessary to prevent erroneous estimation of velocity parallel movement, especially in the period TC.

[0191] the cutoff frequency HPF tunable 2602 and 2603 are intended to reduce the estimated speed parallel motion under the condition in which the speed of the parallel movement becomes very large due to the influence of the acceleration of gravity in connection with the repositioning device and the influence of the perturbing acceleration due to the operation of the pan. Through this it is possible to prevent excessive compensation at compensation of jitter in accordance with the compensation of translational shake in the operation of the pan and changing the position.

[0192] in Addition, in the tenth embodiment, the configuration is applied in which the setting HPF tunable 2602 and 2603 is changed in accordance with the definitions of the pan and change the floor is available. However, the above configuration can also be applied, if the gain of the Kalman changes, as in the first embodiment, and if the input AC gain is changed in the first part of the Kalman filter, as in the eighth embodiment. In this case, by performing the determination of the pan and change the position of the estimated speed of parallel motion in the previous period control can prevent excessive compensation at compensation of jitter in the compensation of translational shake in the operation of the pan and changing the position.

[0193] Although the embodiments of the present invention is described with reference to exemplary embodiments of the exercise, you should understand that the invention is not limited to the disclosed exemplary embodiments of the implementation. The volume of the following claims must comply with the broadest interpretation to encompass all such modifications and equivalent structures and functions.

[0194] This application claims the benefit of patent applications in Japan No. 2011-130614 and 2011-130615, registered on June 10, 2011, which are hereby incorporated in this document by reference.

1. Device for compensation of jitter in the image, configured to compensate for jitter by using the excitation elementarily shake, the device contains:
the compensation unit jitter configured to compensate for jitter by moving not parallel to the optical axis;
excitation unit configured to excite the compensation unit jitter signal-based guidance excitation;
the detection unit position, is configured to detect the position of the compensation unit shake to output a detection signal; and
unit calculating compensation, configured to detect a component of the translational shake in the jitter of the device movement compensation unit jitter, so that the signal specifying the excitation and signal detection positions are used as input, and calculating compensation unit to compensate the jitter of the magnitude of the jitter of the device that includes the component of the translational shake.

2. The device according to p. 1,
in which unit calculating compensation includes surveillance unit containing a model of the vibration compensation unit jitter, so that the signal indications excitation induced block which is the unit compensate for jitter and signal detection position by the block serve as input;
and
in which translational jitter is calculated using the unit of observation.

3. The device according to p. 2,
inwhich the unit of observation is a device evaluation using the Kalman filter that performs a calculation of the translational shake so that the relative displacement between the driven unit and the fixing unit supporting excited by a block which serves as a state variable, and exciting force in the compensation unit shake serves as the input variable.

4. The device according to p. 2, additionally containing:
block positioning, configured to at least determine the status of the pan device or change the location of your device
in which the unit of observation changes the parameters to the calculation result in accordance with the translational jitter was small, if it is determined that there is a state of panning or changing the position of the device, compared with when not determined that there is a state of panning or changing the position of the device.

5. The device according to p. 4,
in which the unit of observation is an estimate using the Kalman filter that performs a calculation of the translational shake so that the relative displacement between the driven unit and the fixing unit supporting excited by a block which serves as a state variable, and exciting force in the compensation unit jitter is used as an input variable; and
in which pileni the Kalman filter is modified as changes in parameters.

6. The device according to p. 2, additionally containing:
the detection unit of angular velocity, is configured to detect the angular velocity of the shake of the device,
in which the computing unit value of compensation contains:
the first block of calculating compensation, configured to calculate the magnitude of the compensating rotational vibration caused due to the rotation of the device, centered on an axis perpendicular to the optical axis of the optical system forming an image using the detection signal of the angular speed detection unit angular velocity; and
the second block of calculating compensation, configured for calculating compensation of translational shake using surveillance unit, so that the relative displacement between the driven unit and the fixing unit supporting excited by a block which serves as a state variable, and excited by a force compensation unit shake serves as the input variable,
in which exciting unit exciting the block compensation of jitter in accordance with the compensation value calculated by synthesizing the value of compensation for the rotational shake and amount of the compensation of translational shake.

7. The device according to p. 6,
in which the second computing unit amount is s compensation includes:
unit calculating a compensation coefficient, configured to calculate the compensation coefficient estimated from the speed of the translational jitter calculated by the surveillance unit, and the detection signal of the angular speed detection unit angular velocity; and
the compensation unit of output, configured to compensate for the output of the first unit calculating compensation using the compensation coefficient from the block coefficient calculation of compensation.

8. The device according to p. 6, further comprising:
the definition block, configured to detect the state of the fluctuations of the angular velocity of the shake device to compare it with a threshold value; and
block determine the gain is configured for changing the gain of the Kalman filter, which is installed previously for the determination of block definitions.

9. The device according to p. 6, further comprising:
the AC gain for the signal specifying the excitation signal or the detection position in the induced block;
the definition block, configured to detect the state of the fluctuations of the angular velocity of the shake device to compare it with a threshold value; and
block determine the gain is configured for changing the gain of the AC gain of the definition wide-angle the block definition.

10. The device according to p. 6, further comprising:
the high-pass filter for the signal specifying the excitation signal or the detection position in the induced block;
the definition block, configured to detect the state of the fluctuations of the angular velocity of the shake device to compare it with a threshold value; and
a setting unit configured to change the cutoff frequency of the high-pass filter on the block definition definition.

11. The device receiving the image, configured to compensate for jitter using an excitation unit compensation shake that contains:
the imaging unit configured to convert the image of the photographic object formed by the optical system forming the image, in the image;
the detection unit position, is configured to detect the position of the compensation unit jitter;
unit calculating compensation, configured to calculate the value of compensation to initiate the compensation unit jitter;
a changing unit configured to change a range of trim images from the imaging unit in accordance with the compensation value; and
excitation unit configured to excite the compensation unit shake,
DG is a block calculating compensation gets the value of the excitation unit compensation shake and the output of the detection unit position, detects the shake of the device, which includes translational jitter caused in connection with the translational movement of the device in the direction perpendicular to the optical axis of the optical system forming the image, and calculates the compensation value; and
where the changing unit changes the range of the cut-off in accordance with the compensation value of the translational jitter calculated by the power calculating compensation.

12. The device according to p. 11, further comprising:
the detection unit of angular velocity, is configured to detect the angular velocity of the shake of the device,
in which the computing unit value of compensation contains:
the first block of calculating compensation, configured to calculate the magnitude of the compensating rotational vibration caused due to the rotation of the device, centered on an axis perpendicular to the optical axis of the optical system forming an image using the detection signal of the angular speed detection unit angular velocity; and
the second block of calculating compensation, configured for calculating compensation of translational shake using surveillance unit, consisting of models of the vibration compensation unit jitter, so that the signal indications excitation vozbujdaet the block, which is the unit compensate for jitter and signal detection position by the block are used as input,
in which exciting unit exciting the block compensation of jitter in accordance with the compensation value of the rotational shake; and
in which the changing unit changes the range of the cut-off in accordance with the compensation value of the translational shake.

13. The method of controlling the compensation of the image vibration performed in the device compensation shake configured to compensate for jitter using an excitation unit compensation shake that contains:
the phase of the excitation, which excite the compensation unit jitter signal-based guidance excitation;
the phase detection position at which detect the position of the compensation unit shake to output a signal detection position;
the step of calculating the amount of the compensation to which detect the component of the translational shake among the shake of the device movement compensation unit jitter, so that the signal specifying the excitation and signal detection positions are used as input, and calculates the compensation value of the compensation unit jitter of the magnitude of the jitter of the device that includes the component of the translational shake.

14. The method of controlling the clipping image running on the device receiving the image, configured to compensate for jitter using an excitation unit compensation shake that contains:
the phase detection position at which detect the position of the compensation unit jitter;
the step of calculating the amount of the compensation to which detect the shake of the device, which includes translational jitter caused in connection with the translational movement of the device in the direction perpendicular to the optical axis of the optical system forming the image, and calculates a compensation value to initiate the compensation unit jitter, using the value of the excitation unit compensate for jitter and signal detection position of the detection step; and
phase changes, which receive the compensation value of the translational shake to change the range of trim images from the imaging unit.



 

Same patents:

FIELD: information technologies.

SUBSTANCE: HPF filter (11) identifies high-frequency components from an input signal of an image Sin. A squaring operator (12) develops a quadratic signal S12 by squaring of the first signal S1. The first differentiating block (13) develops the first differentiated signal S13 by differentiation of the quadratic signal S12. The second differentiating block (14) develops the second differentiated signal S14 by differentiation of the input signal Sin of the image. A multiplier (15) creates the second signal S2 by multiplication of the first differentiated signal S13 by the second differentiated signal S14. The summator (16) develops an output signal Sout of the image as a compensation signal by addition of the second signal S2 to the input signal Sin of the image.

EFFECT: improved quality of an image by sufficient increase of image sharpness by appropriate high-frequency compensation with simple layout, besides, not only for fixed images, but also for moving images, even if images have already been exposed to processes of image enlargement.

21 cl, 28 dwg

FIELD: information technology.

SUBSTANCE: to bring separate single-pixel indicators and separate logic pixels into conformity, a control table is used, which determines the order of displaying image data on the screen of a display device. One single-pixel indicator and one logic pixel correspond to an arbitrary element of the control table. An indicator of a defined colour is activated to generate light flux according to data of the same colour selected from the logic pixel. Selection of these data from the logic pixel and activation of that single-pixel indicator are repeated at high speed. The order of arrangement on the image data plane of geometric centres of non-overlapping or partially overlapping groups from the plurality of logic pixels coincides with the order of arrangement on the surface of the screen of the display device of conditional geometric centres of non-overlapping groups from the plurality of single-pixel indicators, which (groups from a plurality of single-pixel indicators) correspond to those groups from the plurality of logic pixels. Each group from the plurality of logic pixels mutually explicitly corresponds to a group from the plurality of single-pixel indicators. To increase brightness of the image displayed on the screen of the display device, that screen contains white-colour single-pixel indicators which are activated to generate light flux according to data calculated based on data contained in logic pixels which correspond to those white-colour single-pixel indicators.

EFFECT: broader functional capabilities, particularly display of raster data of a colour image on the screen of a display device, higher brightness of the image displayed on that screen.

16 cl, 21 dwg

FIELD: physics.

SUBSTANCE: device comprises a module (101) for detection of a movement vector, which detects a vector of movement in each earlier specified area between frames of the introduced image signal, and a module (2) for identification of circuits, which emphasizes a high-frequency component of the introduced image and a signal of an interpolated image, formed by means of a module (100) for conversion of frame frequency (FRC-module), according to the value of movement of the introduced image signal detected by means of the module (101) for detection of a movement vector. This is compensated by a high-frequency component weakened by means of an image sensor effect of integration in time, to reduce visible blur of moving object images to increase sharpness of a displayed image. By setting an extent (a level) of the interpolated image signal circuits as a lower extent of the signal of the introduced image, the sharpness of the displayed message increases without isolation of the interpolated image signal image deterioration.

EFFECT: displaying video of high sharpness by means of reducing blur of images of moving objects of the displayed video as a result of the image sensor effect of integration in time.

12 cl, 16 dwg

FIELD: processing digital video signals including directional ratio of contours during their assignment.

SUBSTANCE: processing is executed by detection of contour direction relative to point where vertical and horizontal contours are intersecting and relative to diagonal contour point. Times of original pixel and all pixels adjacent to source one in horizontal, vertical, and diagonal directions are coordinated, contour assignment values weighed relative to brightness of pixel having greatest difference in brightness out of adjacent pixels in all directions, and relative to brightness of original pixel are computed and added to original pixel. If two weighted continuous contour assignment values have same sign, preceding and next contour assignment values are received at diagonal contour point, otherwise diagonal optimization is conducted to bring preceding and next contour assignment values to zero, and they are added to original pixel.

EFFECT: more natural processing of inclined contours and contour presenting signals at intersection points of horizontal and vertical contours.

5 cl, 19 dwg

The invention relates to schemes videotrack for a range of videocasts, in particular to schemes underscore circuit

The invention relates to schemes videotrack for a range of videocasts, in particular to schemes underscore circuit

The invention relates to the formation of a digital video signal

The invention relates to the formation of a digital video signal

The invention relates to the formation of a digital video signal

The invention relates to the formation of a digital video signal

FIELD: physics, control.

SUBSTANCE: invention relates to automation devices and can be used in control devices of aerospace equipment. A matrix command generator, comprising a first register (lines), a second register (columns), a first decoder, a second decoder, a first group of switches, a second group of switches, vertical switched buses (columns) and horizontal switched buses (lines), power buses +E and -E of the switched buses, data inputs of the first and second registers are combined and are the data inputs of the device, write inputs of the first and second registers are combined and are the write input of the device; the device also includes a first and a second group of diodes, a matrix command generator, a line fault detector, a column fault detector and a state analyser.

EFFECT: enabling generation of only one command at any moment in time while monitoring working capacity of the generator.

2 cl, 3 dwg, 1 tbl

FIELD: power industry.

SUBSTANCE: invention relates to an optimisation system and method of a technological process for a thermal power plant, and namely to optimisation of load planning of the thermal power plant by using adaptive restrictions. The method and the corresponding system includes detection of an event specifying a necessity of adaptation of one or more restricted target functions used at load planning. With such a detection, a target function is analysed to determine adaptive value of restrictions for one or more restrictions for an optimum solution of the target function. The above adaptive values of restrictions are used to solve the target function, and solution of the target function with one or more adaptive values of restrictions is used to control one or more electric energy generation units of a thermal power plant.

EFFECT: possible minimisation of electric energy generation time.

9 cl, 3 dwg

FIELD: physics, control.

SUBSTANCE: invention relates to method of controlling transmissions of a batteryless device (1) operating in a wireless network. The method comprises steps of: the batteryless device (1) transmitting a frame including elements for controlling operation of a remote device (2a) or controlled device, the batteryless device being configured with a predetermined number of planned retransmissions of the control frame; the batteryless device sensing a change in the physical phenomenon induced by operation of the controlled device (2a); the batteryless device determining, based on the sensing step, the success or failure of the frame transmission; if the transmission has succeeded, the batteryless device omitting further retransmissions of the control frame.

EFFECT: high efficiency of saving power.

14 cl, 1 dwg

FIELD: electricity.

SUBSTANCE: invention suggests a digital electric drive containing a setting device which output is connected to a position regulator with its output connected to a variable-speed control unit, a pulse-width modulator, a servomotor, at that the servomotor is made as a disc toothless thyratron motor divided into three segments and each segment is coupled to an individual power amplifier and a phase current sensor for each segment, which are coupled to a torque valuator, output of the latter is connected to a torque controller, which second input is connected to the output of the variable-speed control unit and output is connected to the pulse-width modulator, the output of the modulator is coupled to three power amplifiers of the motor segments, the electric drive is equipped additionally with a position sensor, which input is connected to the motor and output to the position sensor controller, the output of the controller is connected to the second input of the position regulator and to the second input of the variable-speed control unit through a velocity valuator.

EFFECT: improving reliability index of the digital electric drive system with simultaneous simplification of its assembly and repair procedure due to use of a thyratron motor with a disc stator divided into three segments.

2 cl, 1 dwg

FIELD: physics.

SUBSTANCE: device automatically eliminates static error when used in systems for stabilising dynamic objects by scaling the stimulus. An example of specific implementation of the controller is realised on pneumatic elements of a universal system of elements for industrial pneumatic automation (USEPPA).

EFFECT: high accuracy of control in stabilisation systems with the disclosed device in conditions of both signal and parametric disturbing action on the object; the device includes an adder, an integrator and an amplifier.

4 dwg

FIELD: physics.

SUBSTANCE: illumination system has a light-emitting element which is configured to emit modulated light to a remote control device and detect control signals from the remote control device, which provides communication between the remote control device and the illumination system without an additional sensor or an additional transmitter.

EFFECT: smaller dimensions of the illumination system and design of an illumination system which communicates with a remote control device.

18 cl, 3 dwg

FIELD: physics; control.

SUBSTANCE: invention relates to controllers. Disclosed is a controller, having series-connected control signal source - input terminal, a first adder, an integrator whose output is connected to the input of the first, second and third relay elements, whose outputs are connected to inputs of the first, second and third comparators, respectively; the output of the first, second and third comparators is connected to the D input of a first, second and third dynamic D flip-flops, respectively, phase A, B and C buses connected to the corresponding inputs of a three-phase load with a neutral terminal through the first, second and third power switches, respectively, whose control inputs are respectively connected to the output of the first, second and third dynamic D flip-flop; the controller also has a second adder whose output is connected to the second input of the first adder; fourth, fifth and sixth comparators whose inputs are connected to phase A, B and C buses, respectively; the outputs of the fourth, fifth and sixth comparator are connected to the C input of the first, second and third dynamic D flip-flop, respectively; wherein the controller also has seventh, eighth and ninth comparators, whose outputs are connected to corresponding inputs of the second adder, and the inputs of the seventh, eighth and ninth comparators are connected to outputs of the first, second and third dynamic D flip-flops, respectively.

EFFECT: possibility of self-backup and high reliability of the controller.

8 dwg, 1 tbl

Voltage regulator // 2467376

FIELD: electricity.

SUBSTANCE: voltage regulator is designed for use in power supply systems for regulation, including stabilisation, of single-phase and three-phase voltage of an AC power source, and also for regulation, including stabilisation, of DC voltage. The voltage regulator comprises a computing-metering unit, and at least one regulating unit, at the same time it additionally comprises a tuning unit and a memory unit, besides, the first output of the computing-metering unit is connected to the input of the regulating unit control, and a power input of the regulating unit is connected with an appropriate input of the voltage regulator, a power output of the regulating unit is connected to an output of the voltage regulator, the input and output of the voltage regulator are also connected with inputs of measurement of input and output voltage of the computing-metering unit accordingly, the second of the outputs of the computing-metering unit is connected to the input of the tuning unit, and the output of the tuning unit is connected with the input of the memory unit, besides, the output of the memory unit is connected with an appropriate input of the computing-metering unit.

EFFECT: increased accuracy and reliability of a voltage regulator and simplified design.

3 cl, 7 dwg

FIELD: electricity.

SUBSTANCE: effect is achieved by automatic action on input signals, and device simplification is achieved by the fact that control includes adder, relay module, null element, indicator of extreme points, the input of which is connected to the first input of control and the second input of null element, and information control outputs are connected to the corresponding inputs of relay module the output of which is connected to the first adder input, valuator of average signal value, the information inputs of which are related to information outputs of indicator of extreme points, the control output of which is connected to control input of valuator of average signal value, and output of valuator of average signal value is connected to the first inverse integrator input, the second input of which is connected to the second control input; in addition, adding device is introduced to it, the first input of which is connected to the second control input, and the second input is connected to integrator output, and output of adding device is connected to the second input of adder, and output of null element is connected to control output.

EFFECT: higher control accuracy under conditions of action of relay control signal asymmetry on the object.

3 dwg

FIELD: physics.

SUBSTANCE: method comprises steps for: generating control action and measuring the controlled quantity of the object. Simultaneously with generation at the object of a single control pulse at maximum power, the value of inertial increase in the controlled parameter is constantly calculated, which is used when determining the moment the single control action ends, which enables the controlled quantity to accurately reach the given value over a minimum period of time without overcontrol, and accurate maintenance of the given value of the controlled quantity is carried out by generating calculated power of the control action corresponding to that given value, as well as by constant correction of power of the control action, which takes into account the current control error and the rate of change of the controlled quantity, and constant consideration of properties of the control object during operation.

EFFECT: high efficiency of automatic control of process parameters by taking into account changes in properties of the control object and external disturbances.

4 dwg

FIELD: physics, photography.

SUBSTANCE: present invention relates to an image stabilisation apparatus which corrects image blur (image degradation) caused by shaking, such as hand-shaking, and a control method for said image stabilisation apparatus, an optical apparatus and image forming apparatus. The disclosed group of inventions comprises an optical apparatus and an image forming apparatus, which comprise an image stabilisation apparatus, as well as a control method for the image stabilisation apparatus. The image stabilisation apparatus comprises a vibration correction unit which corrects image blur caused by vibrations; a first vibration detection unit which detects angular velocity of vibrations; a first calculation unit which calculates the value of correcting rotational shaking based on the output signal from the first vibration detection unit; a second vibration detection unit which detects vibrations using a method other than the method of the first vibration detection unit; a second calculation unit which calculates a correction value from output signals of the first vibration detection unit and the second vibration detection unit and calculates the correction value of back-and-forth shaking based on the correction value and output signal of the first vibration detection unit; a drive unit which drives the vibration correction unit based on at least one of the correction value of rotational shaking and the correction value of back-and-forth shaking; and a variation module which, when the main object occupies a smaller portion of the entire screen, changes the correction value of back-and-forth shaking such that it is less than when the main object occupies a larger portion. The disclosed control method for the image stabilisation apparatus comprises steps of detecting angular velocity of vibrations using the first vibration detection unit; calculating the correction value of rotational shaking based on the output signal from the first vibration detection unit; detecting vibrations using the second vibration detection unit using a method other than the method of the first vibration detection unit; calculating a correction value from output signals of the first vibration detection unit and the second vibration detection unit and calculating the correction value of back-and-forth shaking based on the correction value and the output signal of the first vibration detection unit; driving the vibration correction unit based on at least one of the correction value of rotational shaking and the correction value of the back-and-forth shaking; and changing, when the main object occupies a smaller portion of the entire screen, the correction value of back-and-forth shaking such that it is less than when the main object occupies a larger portion.

EFFECT: obtaining images with higher accuracy of stabilisation by correcting vibrations for the main object and correcting vibrations for the background in a reasonable ratio when capturing a moving image.

12 cl, 12 dwg

Up!