Method for video surveillance of open space with fire hazard monitoring
FIELD: physics, video.
SUBSTANCE: invention relates to video surveillance, primarily of open spaces, with fire hazard monitoring and can be used to monitor forest areas in regions with underdeveloped infrastructure. The method comprises video monitoring of a protected area; selecting mobile regions on the obtained image; comparing said regions with reference images from a library of images and making a decision on burning based on the similarity of the obtained images and the available images; wherein mobile objects are selected on the image by breaking down the whole image into rectangular blocks; the shifting of blocks over time is diagnosed from the shift of the centre of a block; separate mobile blocks are merged by spatial clusterisation; time clusterisation of mobile objects is carried out based on a criterion of the intersection of trajectories of separate blocks with each other, wherein the shape of a mobile object is reconstructed by cutting off short edges, joining adjacent boundary points and smoothing the obtained boundaries, wherein the internal region of the object is reconstructed on the obtained boundary, and clusterisation of the selected mobile object is carried out on the normal component of optical flux.
EFFECT: high probability of recognising signs of fire on an open area owing to early detection of fire from smoke traces and air currents.
3 cl, 1 dwg
The invention relates to the field of video surveillance, mostly open space, with the control of fire hazards and can be used to monitor forests in regions with poorly developed infrastructure.
There is a method of event detection for video surveillance systems containing the training phase, in which receive a training image of the monitored area at different points in time in the absence of any subject detection events, and working with the discovery phase, in which the current image mentioned areas, while detect the event by comparing the current image with the image corresponding to a linear combination of a set of reference images that approximate the corresponding training image or coincident with them, on the application number 2009106852/08, publ. 10.09.2010,
The known method is effective for situations in which obviously is no want to monitor events that the control of large open spaces it is difficult to guarantee.
There is a method of surveillance that allows you to control the premises and territory to make the video disturbing events and continuous video recording, according to the patent of Russian Federation №2381533, publ. 10.02.2010,
The method is not sufficiently high information content when the work mode is real time.
There is a method of video surveillance, comprising the steps of receiving a video signal of the controlled area by at least one camera, its transmission and its analysis in at least one computing device memory, which according to the invention obtained by the video signal is analyzed using a computing device using information from descriptions of alarm situations, pre-recorded in a database in the memory of the computing device, based on the analysis results receive the information about the purpose, form a control signal, transmit it to a movable camera, adapted for targeting in accordance with the control signal, using the mobile camera receives an image of the target, suitable for further analysis, and transmit it to store in the database.
Preferably in the method according to the invention according to the results of the analysis additionally generate an alarm signal and transmit it to the operator. The analysis is performed with the opportunity to calculate the velocity vector, the current coordinate location of the target, and optionally generate and transmit mobile video camera corresponding control signal to provide guidance in the mobile camera at the target. The control signal transmitted to the movable video camera via the serial communication channel. Analysis of whitefish is Ala from the camcorder performed with the consideration of objectives, selected from the group including brightness, size, duration of stay in the controlled area, the speed and direction of movement. (Patent RF №2268497, publ. 20.01.2006 was adopted for the prototype of the invention).
There is a method of surveillance and control fire hazards, according to which make the monitoring of the protected area, there is obtained the image of the movable area, compare these areas with exemplary images from a library of images and decide to fire on the basis of similarity obtained and available images.
The technical problem to be solved by the invention is to improve the probability of detection of the signs of a fire in an open area due to the early detection of fire smoke trails and streams of air.
The problem is solved by the fact that in the method of surveillance of open space with the control of fire hazards, according to which make the monitoring of the protected area, there is obtained the image of the movable area, compare these areas with exemplary images from a library of images and decide to fire on the basis of similarity obtained and available images
According to the invention the selection in the image of moving objects is done by splitting the image on to akatie blocks, the blocks from shifting over time, diagnosed according to the displacement of the center of the block, a separate moving blocks combine by spatial clustering temporal clustering of moving blocks produced according to the criterion of intersection of the trajectories of individual blocks among themselves, and form a moving object restore cut-off by short edges, connections lying near the boundary points and smoothing the obtained bounds, while the interior of an object to restore the border, and the classification of the selected movable object produced by the normal component of the optical flow. In addition, a library of images formed with the release categories "vehicle", "people", "animals", "the waving branches of the trees", "smoke", "heat trace", "open fire."
The invention is illustrated in Fig.1, which presents a block diagram of the proposed method.
An example implementation of the method.
1. The selection in the image of the moving blocks
All the image (position 1 in Fig.1) resb is provided on the blocks B isize n×n pixels. The value of n depends on the object size and image resolution, and is selected from the range from 1 to 100. Assume that the center of the block Biis located at the point (x,y) at time t and at the point (x+Δx, y+Δy) at time t+1. Then the magnitude of the matching blocks and their offsets for two consecutive moments of time are expressed as follows:
To eliminate errors of the second kind (false moving blocks), resulting from noise, it is necessary that the value of D has exceeded the threshold δαthat can be determined using a test of significance. For a pixel (x,y), where there is no movement, the value of d(x,y)=It(x,y)-It+1(x,y) represents the noise of the camera. It is known, can be described by a Gaussian distribution with zero mean and variance equal to twice the variance of the noise of the camera. The significance level can be considered as the probability of detecting block as a moving unit, although in reality it is fixed (error of the second kind). If you choose a block size n=16 and put δα=1, then it will fit a probability equal to 0.9 (position 2 in Fig.1).
2. Spatial clustering of moving blocks
At this stage, using clustering techniques are combined separate similar moving blocks, for which the value of D exceeds the threshold value δα, in moving the field. We will consider two blocks are adjacent if they have a common side (or its part) or vertex, i.e. they are related by a 8-adjacency. As characteristic features of BL the Cove, you can use the connectivity C(B iBj), the normalized difference between the matching blocks of D(BiBj) and the normalized block offset Δ(BiBj):
The distance between the moving blocks will be defined as
where the parameter θ is equal to 0.3. Blocks Biand Bjmerged into a single region, if this value is less than threshold δcequal to 0.5 (position 3 in Fig.1).
3. Temporal clustering of moving blocks
The problem with segmentation gaseous objects of the type of smoke is possible nutriwest - at different points in time initially, a single object may consist of several separate areas. In order to identify them as one and the same moving object, will accompany the moving region from frame to frame. Then, while the areas that make up a single object may not touch each other in any frame, the trajectory of these objects will be crossed (position 4 in Fig.1). Thus, the individual objects can be combined in the time domain.
4. Determination of the shape of objects
Obtained in the previous step, the clusters consist of several blocks. To establish the exact shape of the objects used allotment boundary points, for example, by the method of Canny. The obtained edge points for gaseous objects (smoke) are usually not related to each other. To build the closed boundaries of these edges must be applied operations morphology: remove very short edges, as they primarily represent the effect of noise, then use the dilation operation to connect lying near the boundary point, then smooth the resulting borders to give them a more natural appearance.
Let X is a binary the image of the detected edge points. The pixel value equal to 1 means that the pixel is an edge point. The combination of morphological operations to build the boundary of the object from the isolated boundary points as follows:
where ⊕ is the operation of dilatation,
The next step is to restore the internal area of the object obtained by the border. If the area of low intensity are very long, even with the help of morphological operations is impossible to restore the border. This means that the detected boundary is open and may not completely include the interior of an object. Therefore, it is necessary to distinguish areas not encompassed obtained abroad, and independent region. To do this, find the intersection with the boundary in the horizontal and vertical directions. Only if the horizontal and vertical lines passing through the pixel, cross the same border, we assume such a pixel belonging to the internal area of the object (position 5 in Fig.1).
5. Classification of selected objects
Classification of selected objects to determine the smoke and fire is carried out using dynamic textures. One of the main methods of classification using dynamic textures is to use optical flow calculation which is a very demanding problem. We will not determine the full optical flow, but only its normal component. This is sufficient for the purposes of the recognition object type and at the same time will avoid a large amount of computation. Normal flow, i.e. the flow in the direction of the intensity gradient ∇I(x,y) is written as follows:
As characteristic features for the classification of objects in addition to the normal flow will also use the maximum value of the directional regularity textures R(i), which would allow an affine invariance of the results. This approach is based on the fact that the regular (periodic) textures are perceived to be so in a wide range of angles. Therefore, certain adequate, the degree of regularity of the considered objects can serve as invariant, physically sound characteristics. We will use polar coordinates on the grid αi=Δα·i, dj= ∆ D·j.
Determine the maximum value of the directional regularity R(i)
As characteristic features of the textures will use the following averaged over all frames of the sequence of values:
- Average values of
- The ratio of the average values of the normal flow and its standard deviation.
- The degree of uniformity of the orientation of the normal flow
where Ω is the set of pixels where the value of the normal flux is non-zero.
- The average value of P(t).
The variance of P(t).
All these quantities are invariant to translation and rotation. The identity of the object to the class of objects corresponding to the smoke on the image may be determined using any method of classification, such as support vector machines or neural networks (item 6 in Fig.1).
1. Method of surveillance of open space with the control of fire hazards, according to which make the monitoring of the protected area, there is obtained the image of the movable area, compare these areas with exemplary images from the image library and accept solved the e on fire on the basis of similarity obtained and available images, characterized in that the selection in the image of moving objects is done by splitting the image into rectangular blocks, offset blocks, with time, diagnosed according to the displacement of the center of the block, a separate moving blocks combine by spatial clustering temporal clustering of moving blocks produced according to the criterion of intersection of the trajectories of individual blocks among themselves, and form a moving object restore cut-off by short edges, connections lying near the boundary points and smoothing the obtained bounds, while the interior of an object to restore the border, and the classification of the selected movable object produced by the normal component of the optical flow.
2. The method according to p. 1, characterized in that the image library form with the release categories "vehicle", "people", "animals", "the waving branches of the trees", "smoke", "heat trace", "open fire."
3. The method according to PP.1 and 2, characterized in that in each of the categories of image library form a subcategory of "trajectory", "speed", "direction", and bind the resulting images to the characteristic features of the protected area.
FIELD: physics; control.
SUBSTANCE: invention relates to video monitoring systems and a control method therefor. The video monitoring system comprises a signal interface gateway which receives, respectively, control signals from a digital video monitoring system and/or a matrix video system in an analogue video monitoring system and accordingly converts the control signals from the digital video monitoring system and/or control signals from the matrix video system. Said conversion includes creating a relationship table in the signal interface gateway which represents relationships between channels of the digital video monitoring system and channels of the matrix video system, and performs conversion of the control signals according to the relationship table.
EFFECT: providing mutual control of an analogue video monitoring system and a digital video monitoring system.
6 cl, 10 dwg
FIELD: physics, signalling.
SUBSTANCE: invention relates to security signalling, particularly, to video surveillance means for detection and identification of trespassers that penetrate through security borders and actuate the detection means. Proposed system consists of central control board and sets of remotely addressed video cameras connected with central control board via switchboard channel. Switchboard channel is composed by digital data transmission network via two-way serial communication interface.
EFFECT: higher reliability owing to higher quality of video pictures, around-the-clock video surveillance with IR-floodlight illumination of video camera coverage area.
13 cl, 5 dwg
FIELD: physics, video.
SUBSTANCE: invention relates to video surveillance systems using means of recognising hazardous at a secure facility. The device comprises at least one video camera which is capable of converting a video signal to a digital video signal, a video signal storage unit, a unit for converting the digital video signal to a sequence of images on a video monitor screen, and further includes a manipulator of a control unit of the conversion unit which is such that movement of the manipulator corresponds to change in images on the video monitor screen, and the direction and rotational speed of the wheel corresponds to the sequence and rate of change of successive images; the device includes an event analyser which picks up events from the digital video signal which require the attention of the operator, and generates a labelling signal upon picking up an event which requires the attention of the operator; the device further includes an event display in form of a wheel, which is such that its rotation corresponds to the rotation of the manipulator of the control unit, and the event display displays images of labels of events recorded by the camera which require the attention of the operator.
EFFECT: high reliability of security and accuracy of recognition.
8 cl, 3 dwg
FIELD: radio engineering, communication.
SUBSTANCE: playback control system comprises: a video recording means, a motion detecting means and a video playback means, wherein the video recording means is capable of recording video surveillance; the motion detecting means is capable of recognising a video image in real time and labelling a time index for a dynamic frame (dynamic frames) of the video when recording using the video recording means; and the video playback means is capable of obtaining the time index from the motion detecting means and playback the dynamic frame(s) of the video surveillance record in accordance with the time index when playing back the video surveillance record.
EFFECT: shorter time for playing back a video surveillance record without loss of significant information.
8 cl, 2 dwg
SUBSTANCE: computer includes a video card; in the television camera, a first television signal sensor (TSS) is based on an array of charge-coupled devices with "row-frame transfer", and the use of an additional pulse former for clock powering of the photodetector enables summation of charge signals in the memory section accumulated in the photodetector section thereof. As a result, the signal-to-noise ratio for a general view image increases in proportion to the increase in energy of the useful signal thereof, i.e. the number of accumulation frames.
EFFECT: high quality of the general view image due to a higher signal-to-noise ratio at the output of the array of charge coupled devices of the first television signal sensor owing to summation, in the memory section thereof, of charge packets formed in the photodetector section.
5 cl, 10 dwg, 2 tbl
FIELD: information technology.
SUBSTANCE: method includes a setup step where an operator creates a model of the environment by determining the number of cells corresponding to regions of said environment, and then creates a "cell/sensor" connection by defining for each sensor at least one possible position associated with at least one cell. For each position, the operator assigns the sensor a control estimate for the associated cell. The method also includes a working step where a control system, in order to execute a surveillance function, finds those sensors that can be used to execute the requested surveillance function and control said sensors based on control estimates and the "cell/sensor" connections.
EFFECT: high reliability of surveillance systems, particularly video surveillance systems, by providing dynamic adaptation each time a sensor cannot be used by the system.
9 cl, 6 dwg
FIELD: information technology.
SUBSTANCE: passenger carriage video monitoring system has at least one video camera mounted such that it is capable of scanning at least part of the interior of the passenger carriage. The video camera is connected to a video recorder to which a monitoring and control unit is reversibly connected. The video recorder can be connected to a data transmission system and a monitor mounted in the guard's compartment.
EFFECT: safer transportation of passengers.
FIELD: information technology.
SUBSTANCE: apparatus for detecting and monitoring hazard with a built-in display system is meant for use in a public safety system and warning on emergency situations, adapted for detecting and identifying hazards in the surrounding medium, as well as displaying information of interest relating to the hazard or other public information.
EFFECT: detecting hazard and endangering activities at early stages thereof.
25 cl, 3 dwg
SUBSTANCE: system has a sensor unit 1, a signalling unit 2, a video surveillance unit 3, an actuating device unit 4, first 5 and second 6 modems, an information processing and display unit 7, a radio channel 8, a switching unit 9, a video information superposition unit 10, a display unit 11, audio signal transmitting 12 and receiving 13 units, a video peephole 14, a video signal converter 15 and a low frequency amplifier with a microphone 16 (i=1, 2,…, n). The first modem 5 has driving generators 17.i, a first OR logic element 18, a modulating function former 19, an amplitude modulator 20, a modulating code former 21, a phase-shift modulator 22, a power amplifier 23 and a transmitting antenna 24.
EFFECT: high reliability of security and surveillance of residential facilities and other secure objects through use of a radio channel using signals with combined amplitude modulation and phase-shift modulation.
FIELD: information technology.
SUBSTANCE: system is equipped with two memory modules, two analysis and recognition units, one having a function for analysing and comparing records from the first frame to the last, and the other having a function for analysing and comparing records from the last frame to the first, and a second video monitor, where each television camera is connected through a data network to the corresponding analysis and recognition unit which is connected to the corresponding memory module and the video monitor. Television cameras are arranged such that they can be simultaneously rotated in the horizontal plane by a rotary drive. Optical axes of the television cameras lies on a straight line and are directed oppositely. Motion detectors lie on the security perimeter. The output of each motion detector is connected to the corresponding pulse former, whose output is connected to the input of an OR element, whose output is connected to control inputs of the television cameras and the rotary drive. Efficiency of the security television system lies in elimination of blind spots inside the perimeter of the secured object, which prevents intrusion of the facility.
EFFECT: high efficiency of inspecting a secure territory on the perimeter and avoiding intrusion, achieved due to that the security television system comprises two television cameras connected through a data network to a video monitor and motion detectors connected to a signalling unit.
FIELD: video surveillance.
SUBSTANCE: method includes video surveillance of controlled object state, while into television cable main of object high-frequency television modulated signal is sent, while to receive signal concerning state of S objects, each of which includes group of N video surveillance blocks, including camera and microphone, video-audio signals from each group of N video surveillance blocks are combined along low frequency, received complex video signal is converted from each group of N video surveillance blocks into high-frequency television modulated signal and it is synchronized with unified cable main - coaxial television cable, in arbitrary groups combination, via which received independent S signals are sent to input of visualization and/or recording systems.
EFFECT: higher efficiency.
3 cl, 2 dwg
FIELD: video surveillance technologies.
SUBSTANCE: method includes surveillance of state of object, by surveillance blocks, each of which includes camera and microphone, low-frequency signals received from each block are converted to high-frequency television modulated signals, which are inputted into unified cable main, formed by coaxial television cable, along which received independent signals are sent to input of control panel, in form of television receiver or computer, provided with extension board, allowing to receive and display images of several surveillance objects concurrently, while power is fed along coaxial chamber of television cable main.
EFFECT: higher efficiency.
2 cl, 3 dwg
FIELD: railway transport; signaling and communication.
SUBSTANCE: proposed system contains U-shaped carrying installed over track. Optoelectronic car counter, train beginning pickup, wheel counter, video camera of car left side and video camera of car right side are mounted on vertical supports of carrying structure. Video camera of car roof is secured on carrying structure crossmember and searchlights are mounted on vertical supports and on crossmember. Outputs of pickups are coupled through indicator and matching unit with input of workstation system unit. Outputs of video cameras are coupled with corresponding video inputs of workstation system unit. Workstation is made for displaying image of passing train on computer monitor screen in real time simultaneously from three cameras under POLYSCREEN conditions, and also for full-screen viewing of train from one of chosen video cameras, recording of compressed video from three video cameras of hard disk, reproducing of video record and forming of video archive.
EFFECT: improved on-line detection of commercial troubles on rolling stock.
FIELD: property protection technologies.
SUBSTANCE: alarm signaling device is used to detect an attempt of unsanctioned action against protected object and appropriate alarm message is generated with identification signs and location parameters of current alarm signaling device. Then by radio alarm message is sent to adjacent alarm signaling device, and relayed further to other alarm signaling devices until received by user equipment. On receipt identification signs and position parameters of source alarm signaling device are selected. Using said position parameters flight route is set and unmanned aircraft is launched, made with possible horizontal movement according to set program, video surveillance of segments of protected object, compression and transmission of received video frames by radio to user equipment. Previously, in training mode, launches of unmanned aircraft are performed along radial routes under route angles, matching shortest distances from user equipment to alarm signaling devices position points, and series of received video frames are recorded to memory of onboard complex of aircraft. When receiving an alarm message from one of alarm signaling devices commands for controlling flight of aircraft to it along set route and commands for stabilizing it according to flight parameters are formed by combined correlation processing onboard the aircraft of current and prerecorded video image frames, appropriate for radial flight route of unmanned aircraft to current alarm signaling device.
EFFECT: higher efficiency, broader functional capabilities.
6 cl, 6 dwg
FIELD: television cameras, working under conditions of complex lighting and/or complex brightness of objects, when simultaneously strongly and weakly lit objects and/or objects with sharp difference in brightness may be in camera field of view.
SUBSTANCE: components of combined image are formed in charge form on targets of first and second television sensors with various exposition times, optimal for each one of transferred scene fragments. Due to that, in output image of television camera, bright and light parts are transferred same as in prototype, without limitation of white, while dark and/or low light parts are transferred with high signal/noise ratio, and dynamic range of brightness degrees is therefore extended.
EFFECT: extended dynamic range of brightness degrees for control objects, transferred in combined image outside the "window" due to increase of signal/noise ratio for dark and/or low light parts of objects.
6 cl, 4 dwg
FIELD: analysis of television images, possible use in video surveillance systems.
SUBSTANCE: in accordance to the invention, a set of digital video data from first source, representing first image, is identified as standard video data of first source. Then a second set of video data is read, which represents current image. Difference ratio is computed using standard digital video data and current set of digital video data. If difference ratio exceeds a threshold, a query is shown to system user on the display to receive a response about whether current digital video data belong to identified source, or originate from new source. If response points at new source, then current set of digital video data is dispatched for storage into second memory cell, connected to second source. Then the current set of digital video data is identified as standard digital video data for second source.
EFFECT: division of digital video data outputted by several sources.
5 cl, 9 dwg
FIELD: railway transport.
SUBSTANCE: invention relates to remote control of sizes of loads carried by rail. Method includes checking of observation of obstruction clearance limits, when train enters check zone, by means of optoelectronic sensors and determining whether loads lie within gauge limits when train passes check zone formed by optical limits of clearance gate sensors, comparing loading gauges with required train gauges and video observation of passing train. Set of two pairs of optoelectronic sensors "radiator-receiver" are installed in each plane under checking, time of overlapping of optical channel is measured as well as speed of train, and length violation of gauge is calculated. At least two radiation receivers are installed in plane under checking at a distance, and radiators mated with receivers are installed close to each other as possible to form optical axes arranged at angle relative to each other. Out-of-gauge coordinates are determined and obtained information is transmitted by communication channels, being visualized on screen of videomonitor, and attention of operator is drawn to violation, alarm signal is sent and response of operator is recorded. Device contains speed meter, radiators, receivers, computing device, videomonitor and operator response detector.
EFFECT: improved reliability.
3 cl, 1 dwg
FIELD: observation of moving objects.
SUBSTANCE: method includes using movement sensors, capable of recording two-dimensional distributions of intensity in form of images, where sensors are positioned with known spatial orientation, making it possible to perform simultaneous observation of one and the same scene, periodical query of sensors is performed during whole time period after their enabling, processing and analysis of data received from sensors is performed, which constitutes series of images, and output signal is generated in case of detection of three-dimensional moving object and determining of its spatial position, which signal is injected into output device.
EFFECT: increased trustworthiness when determining spatial position of a moving object.
3 cl, 1 dwg
FIELD: physics, video technology.
SUBSTANCE: camera includes image sensor module, suitable objective lens, particularly wide-angle lens, and holder. The lens features threading which, via the holder and elastic spacing piece, is screwed to match thread of image sensor, so that the thread performs two functions, namely function of image sensor module connection by the lens, and function of focusing by adjusting optic distance between the image sensor module and the lens. Lens thread passes through traverse holes in the holder and elastic spacing piece in the form of elastic ring positioned between the lens and holder. Various vision directions of image sensor module can be adjusted by rotation element.
EFFECT: camera suitable for house intercom, with more flat and compact design and improved functional capability.
22 cl, 12 dwg
SUBSTANCE: method of determining liquid cargo weight in railway tanks consists in measuring liquid cargo level by thermovision image of tank lateral surface and calculating cargo volume and weight. It differs from known methods in that, natural sheet or shipping documents for the tank are used to determine tank type, declared cargo type and weight. Tank lateral surface part thermovision fragment corresponding to filled tank section is used to determine filled cargo temperature, and liquid cargo weight is calculated with due allowance for declared type and specific weight at measured temperature and tank type. Tank is weighed to subtract empty tank weight from measured weight to determine actual weight of liquid cargo. Now, said weight is compared with declared weight to conclude on commercial safety of cargo.
EFFECT: higher validity of cargo commercial safety control.
2 cl, 1 dwg