Method, apparatus and computer programme product for compensating eye colour defects

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to means of compensating for eye colour defects on an image. The method includes computing a first difference image based on the difference between red pixel intensity and green pixel intensity of each set of pixels associated with a first eye region on the image, processing the first difference image to compute a gradient and a projection associated with said gradient, determining a central point based on the projection and computing a first eye colour defect region based on the central point and the plurality of red pixels from said set.

EFFECT: high efficiency of detecting red eye effect on an image.

25 cl, 8 dwg

 

Area of technology

Various embodiments of the present invention generally relate to a method, device and computer program product for compensating eye color defects.

The level of technology

Defects eye color (e.g., red eye) is usually seen on images obtained with a low level of ambient light. As a rule, in low light the flash is performed by the capture device image (e.g., camera), is reflected by the pupils of the eyes of the subject, which often results in an unnatural coloration to the area of the pupil in the image in red. In some cases, the subject's eye can be painted in other reddish colors like Golden, orange, etc. the occurrence of the phenomenon of the eye color defect may also refer to various other factors, such as the small angle between the flash lamp and the lens device of image capture, the proximity of the subject to the capture device image, etc.

As a rule, to avoid the appearance of red eyes in images, we use the capture device image, which may contain a secondary lamp or light source, performing a flash to capture images. Secondary lamp shines in the eye of the subject for a few seconds before the actual sun�the cap, giving pupils sufficient time to their constriction. In some cases, instead of one full of flash device image capture provides many pre-flashes, which can have the same effect as single flash secondary lamp. However, such device can introduce a delay between the first outbreak and the actual capture of the image, whereby the subject may deviate from the preset position before will be captured actual image.

In some cases, for the correction of eye color defects in the images may require user intervention, which is to process the captured image to remove color defects of the eye. However, such user intervention, consisting in the identification and elimination of defects the eye color, can be burdensome and inefficient. In addition, such devices can often miss other well-known defects of the eye color other shades and brightness, for example, Golden eyes, orange eyes, etc.

Summary of the invention

Various aspects of exemplary embodiments of the invention described in the claims.

In the first aspect of the invention provides a method, comprising: calculating a first difference image based on the difference between artistocracy pixels and brightness of the green pixels from the set of pixels, wherein said set of pixels associated with the first region of the eye in the image; processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient; determining at least one Central point based on the specified at least one projection and calculation of the first region of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

In a second aspect of the invention provides a method, comprising: defining a first region of the eye color defect in the low-resolution image (LR, low resolution), while the low-resolution image formed from the high-resolution image (HR, high resolution; determining information about the eye color defect for low resolution images, wherein said defects of eye colors contains a set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image; convert the information about the eye color defect, calculated on the basis of low resolution images, so, to make it suitable for use on high-resolution image, while you provide during account creation�e high-resolution image contains the second area of the eye color defect, corresponding to the first region of the eye color defect in the low-resolution image, and a lot of second peripheral portion of the second region of the eye color defect corresponding to the set of the first peripheral portion of the first region of the eye color defect, and processing high-resolution images to identify the eye color defect in at least one of the plurality of second peripheral portion based on a specified set of identifiers.

In the third aspect of the invention proposes a device comprising at least one processor and at least one memory containing computer program code, wherein the specified at least one memory and computer program code configured to, with the specified at least one processor to enforce the specified device at least the following: computing a first difference image based on the difference between the luminance of the red pixel and the brightness of the green pixels from the set of pixels, wherein said set of pixels associated with the first region image; processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient; identifying at least od�Oh center point based on the specified at least one projection and calculation of the first region of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified the set of pixels.

In a fourth aspect of the invention proposes a device comprising at least one processor and at least one memory containing computer program code, wherein the specified at least one memory and computer program code configured to, with the specified at least one processor to enforce the specified device at least the following: determine the first region of the eye color defect in the low-resolution image (LR, low resolution), the image low resolution image formed from a high-resolution (MR, high resolution); determining information about the eye color defect for low resolution images, wherein said defects of eye colors contains a set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image; converting information on the eye color defect, calculated on the basis of low resolution images, so as to make it suitable for use on high-resolution image, wherein the said high-resolution image contains the second area of the eye color defect corresponding to the first region of the eye color defect in the low-resolution image, and in many�contain peripheral portion of the second region of the eye color defect, corresponding to the multiple first peripheral portion of the first region of the eye color defect, and processing high-resolution images to identify the eye color defect in at least one of the plurality of second peripheral portion based on a specified set of identifiers.

In the fifth aspect of the invention proposes a computer program product comprising at least one computer-readable storage medium containing a set of instructions that when performed by one or more processors provide performance of the device at least the following: computing a first difference image based on the difference between the luminance of the red pixel and the brightness of the green pixels from the set of pixels, wherein said set of pixels associated with the first region of the eye in the image; processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient; identifying at least one Central point based on the specified at least one projection and calculation of the first region of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

In the sixth aspect of the invention proposed�. a computer software product, comprising at least one computer-readable storage medium containing a set of instructions that when performed by one or more processors provide performance of the device at least the following: determine the first region of the eye color defect in the image is low resolution, the low resolution image formed from the high-resolution image; determining information about the eye color defect for low resolution images, and the information about the eye color defect contains a set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image; converting information on the eye color defect, calculated on the basis of low resolution images, so as to make it suitable for use on high-resolution image, wherein the said high-resolution image contains the second area of the eye color defect corresponding to the first region of the eye color defect in the low-resolution image, and a lot of second peripheral portion of the second region of the eye color defect corresponding to the set of the first peripheral portion of the first region of the eye color defect, and processing high-resolution images to identify the eye color defect in at least �the bottom of the plurality of second peripheral portion based on a specified set of identifiers.

In the seventh aspect of the invention proposes a device comprising: means for computing a first difference image based on the difference between the luminance of the red pixel and the brightness of the green pixels from the set of pixels, wherein said set of pixels associated with the first region of the eye in the image; means for processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient; means for identifying at least one Central point based on the specified at least one projection and a means for calculating a first region of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

In the eighth aspect of the invention proposes a device comprising: means for determining a first region of the eye color defect in the image is low resolution, the low resolution image formed from the high-resolution image; means for determining the set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image, means for converting the information on the defect color GLA� for low resolution images, so, to make it suitable for use on high-resolution image, wherein the said high-resolution image contains the second area of the eye color defect corresponding to the first region of the eye color defect in the low-resolution image, and a lot of second peripheral portion of the second region of the eye color defect corresponding to the set of the first peripheral portion of the first region of the eye color defect, and means for processing the high-resolution image to identify the eye color defect in at least one of the plurality of second peripheral portion based on a specified set of identifiers.

In the ninth aspect of the invention provides a method, comprising: calculating a first difference image based on the squared difference between the luminance of the red pixel and the maximum value of the luminance of the green pixel and blue pixel from the set of pixels, wherein said set of pixels associated with the first region of the eye in the image; processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient; determining at least one Central point based on the specified at least one projection and calculation of the first region �effect of eye color on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

In the tenth aspect of the invention proposes a device comprising at least one processor and at least one memory containing computer program code, wherein the specified at least one memory and computer program code configured to, with the specified at least one processor to enforce the specified device at least the following: computing a first difference image based on the squared difference between the luminance of the red pixel and the maximum value of the green pixel and blue pixel from the set of pixels, wherein said set of pixels associated with the first region image; processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient; identifying at least one Central point based on the specified at least one projection and calculation of the first region of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

In the eleventh aspect of the invention proposes a computer program containing a set of instructions that when performed by one or more proce�the Sora ensure the fulfillment of the device at least the following: computing a first difference image based on the difference between the luminance of the red pixel and the brightness of the green pixels from the set of pixels, wherein said set of pixels associated with the first region of the eye in the image; processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient; identifying at least one Central point based on the specified at least one projection and calculation of the first region of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

Brief description of the drawings

For a more complete understanding of embodiments of the present invention is below described with reference to the accompanying drawings.

Fig.1 schematically shows a digital image illustrating the eye color defect according to the exemplary embodiment of the present invention.

Fig.2 shows the device according to the exemplary embodiment of the present invention.

Fig.3 shows another device according to the exemplary embodiment of the present invention.

Fig.4A and 4B shows a horizontal projection and vertical projection associated with at least one gradient according to the exemplary embodiment of the present invention.

Fig.5A and 5B shows a horizontal projection and vertical projection, associerad�nnaya at least one gradient according to another exemplary embodiment of the present invention.

Fig.6 shows a diagram illustrating the color defects of the eye in an image of low resolution (LP) and the corresponding high-resolution image (HP) according to the exemplary embodiment of the present invention.

Fig.7 shows the sequence of operations illustrating a method of compensating eye color defects in the images according to the exemplary embodiment of the present invention.

Fig.8 shows the sequence of operations illustrating a method of compensating eye color defects in the images according to another exemplary embodiment of the present invention.

Detailed description of the invention

The exemplary embodiments of the present invention and their potential advantages described with reference to Fig.1-8.

Fig.1 schematically shows a digital image illustrating the eye color defect according to the exemplary embodiment of the present invention. Fig.1, the image 100 shows a human face 102. The face 102 contains the eye, such as eye 104. Each eye contains the pupil, iris and sclera. It is assumed that the eyes in a human face 102 is exposed to the phenomenon of the eye color defect, and accordingly the pupil 106 may have one of the shades of red, for example, a rich red color, Golden color, orange color etc. To illustrate the phenomenon of the eye color defect Fig.1 the pupil 106 eyes 04 light shows, and not black. The phenomenon of the eye color defect is also known as "red eyes", and in the present description the terms "red eye" and "eye color defect" can be used interchangeably.

As described above, the image 100 may be captured using the capture device image, such as the camera. Other examples of capture device image may include, without limitation, film camera, digital camera, camcorder, camera, mobile phone or any other multimedia device with the features of image capture. The camera can capture the image using configured it flashes, which can cause in the captured image of the subject color defects of the eye, as shown in Fig.1. As stated above, the image captured by the camera can be analyzed to detect defects eye color, and then this effect can be compensated. The detection of the eye color defect may be performed by the capture device image, or alternatively, the captured image may be transmitted to the computing device that is capable of detecting color defects of the eye. Such methods, devices and computer program product for compensating eye color defects described in detail with reference to Fig.1-8.

Fig.2 shows an example of a device�200 according to an embodiment of the present invention. However, it is clear that the device 200 illustrates just one of the types of devices that can be implemented and the advantages of different embodiments of the present invention, and therefore the device does not limit the present invention. Moreover, it is clear that at least some of the components described below in connection with the device 200 may be optional, and thus, an example embodiment of the present invention may include more components, fewer components, or other components than those described in the embodiment of the present invention shown in Fig.2. The device 200 shown in Fig.2, may be any of a variety of types of mobile electronic devices, such as portable digital Secretary (PDA), mobile television, gaming device, cell phone, computer of any type (e.g., a laptop, a mobile computer or desktop computer), camera, media player, mobile digital Secretary or any combination of these and other types of multimedia devices.

The device 200 may include an antenna 202 (or multiple antennas), functionally connected to the transmitter 204 and receiver 206. The device 200 may also include another device such as a controller 208 or other device �of processing, which issues signals to the transmitter 204 and receives signals from the receiver 206. The signals may include signaling information corresponding to the wireless interface standards of the applicable cellular system, and/or may contain data corresponding to speech, received data and/or data generated by the user. In this regard, the device 200 can be configured to work in accordance with one or more standards interface wireless, communication protocols, modulation types, and access types. For example, device 200 may operate in accordance with any of a variety of communication protocols of the first, second, third and/or fourth generation, etc. for Example, device 200 may operate in accordance with wireless communication protocols for second generation (2G) is-136 (multiple access with time division multiplexing (TDMA, Time Division Multiple Access), GSM (Global System for Mobile Communications, global system for mobile communications) and IS-95 (multiple access code division multiplexing (CDMA, Code Division Multiple Access)), or protocols of wireless communication of the third generation (3G) networks such as universal mobile telecommunications system (UMTS, Universal Mobile Telecommunications System), CDMA2000, wideband CDMA (WCDMA, Wideband CDMA synchronous CDMA system with time division of channels (TD-SCDMA, Time Division-Synchronous CDMA), with protocols 3.9 G BES�roodney communication for example-developed universal terrestrial radio access network (E-UTRAN, Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network), wireless communications protocols the fourth generation (4G), etc. Alternative or additionally, the device 200 may operate in accordance with the mechanisms nestaway context, for example, in computer networks such as the Internet, local network, global network, etc.; network wireless short-range communication such as Bluetooth®, Zigbee®, IEEE network 802.Them, etc., wired telecommunications network such as public switched telephone network (PSTN, Public Switched Telephone Network).

The controller 208 may contain electrical circuits that implement, among other things, the processing of the audio and logic functions of the device 200. For example, the controller 208 may include, without limitation, one or more devices of digital processing of signals, one or more microprocessors, one or more processors with the accompanying processor (CPUs), digital signal, one or more CPUs without supporting CPU (processor), a digital signal, one or more specialized computer chips, one or more programmable gate arrays (FPGA, Field Programmable Gate Array), one or more controllers, one or more application specific integrated circuits (ASIC, pplication Specific Integration Circuit), one or more computers, various analog-to-digital converters, digital-to-analog converters and/or other support circuits. Control functions and signal processing device 200 is allocated between these devices according to their functional capabilities. The controller 208 may include functions convolutional encode and interleave message and data prior to modulation and transmission. The controller 208 may additionally include an internal voice coder, and an internal modem for data transmission. In addition, the controller 208 may include functionality to perform one or more programs that may be stored in memory. For example, the controller 208 may be capable of communication software, such as a conventional web browser. The program may allow the device 200 to transmit and receive web content, such as content-based location and/or other web pages in accordance with, for example, WAP (Wireless Application Protocol, an application Protocol for wireless communication), HTTP (Hypertext Transfer Protocol, hypertext transfer Protocol) and/or similar protocols. In the exemplary embodiment of the present invention, the controller 208 may be implemented as a multi-core processor like dual core or curreny processor. The controller 208 may contain any number �of recession.

The device 200 may also contain a user interface that includes an output device, such as a call 210, earphone or speaker 212, a microphone 214, a display 216 and a user interface input that can be connected to the controller 208. User input interface, which allows the device 200 to receive data, may include any of a number of devices allowing the mobile terminal 10 to receive data, for example, a keyboard 218, a touchscreen or other input device. In embodiments including a keypad 218, a keyboard 218 may include a standard numeric keys (0-9) and related keys (#, *), and other keys used to control device 200. In an alternative embodiment, the keypad 218 may be configured as a standard QWERTY keyboard. Keyboard 218 may also include various soft keys with associated functions. In addition or in an alternative embodiment, the device 200 may include an interface device such as a joystick or other user input interface. The device 200 also may include a battery 220 such as a vibrating battery, which is used to power various circuits necessary for the operation of the device 200, and also provides mechanical vibration as Ho�Osho distinguishable output signal.

In the exemplary embodiment of the present invention, the device 200 contains an element for capturing images, such as camera, video and/or audio module associated with the controller 208. Element for capturing the image can be any means for capturing images, and to store, display or transmission of the image. In an embodiment of the present invention, in which the element for image capture is a module 222 of the camera, the module 222 of the camera may include a digital camera capable of forming from the captured image a digital image file. The module 222 of the camera contains all of the hardware, such as a lens or other optical component (components), and software necessary to create a digital image file from a captured image. Alternatively, the module 222 may include only the hardware needed to view an image, while the storage device in the device 200 can be stored in the form of software instructions for execution by the controller 208, providing the creation of a digital image file from a captured image. In yet another embodiment of the present invention, the module 222 of the camera may also include a processing element, such as a coprocessor, which�oget controller 208 to process the image data, both the encoder and/or decoder for compressing and/or decompressing image data. The encoder and/or decoder may encode and/or decode according to the standard JPEG format or other similar format. For the video encoder and/or decoder may use any of a variety of standard formats, such as standards N. 261, N.262/MPEG-2, H. 263, N.264, N.264/MPEG-4, MPEG-4, etc. In some cases, the module 222 may provide the display 216 of the image data in real time. In addition, in one embodiment of the present invention, the display 216 may be located on one side of the device 200, and the module 222 of the camera may include a lens mounted on the opposite side of the device 200 on the display panel 216 that allows the module 222 of the camera to capture images on one side of the device 200 and display such images to the user on the other side of the device 200.

The device 200 may also contain module 224 user identification (UIM user identity module). Module 224 user identification may be a storage device with an embedded processor. Module 224 user identification may, for example, contain a subscriber identification module (SIM, Subscriber Identify Module), built-in universal card (UICC, Universal Integrated Circuit Card), universal, combined with led module�ation subscriber (USIM, A Universal Subscriber Identify Module), plug-in module user identification (R-UIM, Removable User Identify Module) or any other smart card. Module 224 identify the user typically stores information elements related to a mobile subscriber. In addition to the module 224 user identification device 200 may be equipped with memory. For example, device 200 may include volatile memory 226, such as a volatile memory with random access (RAM) including a cache memory for temporary storage of data. The device 200 may also include non-volatile memory 228, which may be embedded and/or removable. Non-volatile memory 228 may additionally or alternatively include an electrically erasable programmable read only memory (EEPROM), flash memory, hard disk, etc., the Memory can store any of a variety of elements of information and data used by the device 200 to implement the functions of the device 200.

Fig.3 shows a device 300 according to one embodiment of the present invention. The device 300 can be used, for example, in device 200, shown in Fig.2. However, it should be noted that the device 200 can also be used in many other devices, both mobile and stationary, and therefore, embodiments of the present invention is not Ogre�limited to the use of the device such as device 200 shown in Fig.2. Alternatively, embodiments of the present invention can be used in combination of devices including, for example, the aforementioned device. Accordingly, various embodiments of the present invention can be implemented entirely in the device (e.g., device 200, or by devices associated with respect to client-server). In addition, it should be noted that the devices or elements described below are not mandatory, and thus, some of them in certain embodiments of the present invention can be omitted.

In one embodiment of the present invention, the device 300 provides compensation of eye color defects. In one embodiment of the present invention, the device 300 may be referred to as device 300 for compensating eye color defects. The device 300 comprises at least one processor 302, the at least one memory 304, user interface 306, an interface 308 for communication and a decoder 310, or otherwise associated with them. Examples of the at least one memory 304 include, but are not limited to, volatile and/or nonvolatile memory. Some examples of volatile memory include, but are not limited to, memory random access, dynamic amp�th random access, static memory with random access, etc. Some examples of non-volatile memory include, but are not limited to, hard disks, magnetic tapes, optical disks, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable permanent memory, flash memory, etc., the Memory 304 may store information, data, applications, instructions, etc. that allows the device 300 to perform various functions in accordance with various embodiments of the present invention. For example, memory 304 may perform the buffering of input data for processing by processor 302. Additionally or alternatively, the memory 304 may store instructions for execution by processor 302.

The processor 302, which may represent an example of the controller 208 shown in Fig.2 may be implemented in various ways. The processor 302 may be implemented as multi-processor, single-core processor, or a combination of multi-core processors and single core processors. For example, the processor 302 may be implemented as one or more of various processing means such as a coprocessor, a microprocessor, controller, digital signal processor (DSP), processing circuitry with an accompanying DSP, or various other processing device�, including integrated circuits, for example, a specialized integrated circuit (ASIC), field programmable gate arrays (FPGA), a microcontroller unit (MCU), a hardware accelerator, computer chip special purpose, etc. In an embodiment of the present invention, the multicore processor may execute instructions stored in memory 304 or accessible by the processor 302 in a different way. Alternative or additionally, the processor 302 may be configured for execution by a hardware-implemented functions. Regardless of whether the configured processor by means of hardware or software or combinations thereof, the processor 302 is an object, for example, physically implemented in a circuit, capable if configured to perform operations according to different variants of implementation of the present invention. Thus, for example, when the processor 302 comprises two or more circuits ASIC, FPGA, etc., the processor 302 may be specifically configured hardware for conducting the operations described above. Alternatively, in another example, when the processor 302 is made in the form of means of execution of program instructions, the instructions when performed can configure the processor 302 to perform the algorithms described above and/ilioperato. However, in some cases, the processor 302 may be a processor of a specific device such as a mobile terminal or network device, adapted to implement embodiments of the present invention by further configuration of the processor 302 using instructions that enable you to perform the algorithms and/or operations described above. The processor 302 may contain, among other things, a clock, an arithmetic logic unit (ALU) and logic gates designed to support the operation of the processor 302.

User interface 306 may be coupled to the processor 302. Examples of the user interface 306 include, but are not limited to, the interface input and/or output. Input interface configured to receive an indication of user input. User output interface can provide the user with audible, visual, mechanical or other output and/or feedback. Examples of the input interface may include, without limitation, a keyboard, mouse, joystick, keyboard, touch screen, soft keys, etc. examples of the output interface may include, without limitation, a display, such as led display, thin-film transistor (TFT), liquid crystal display, active-matrix display based on organic light�diodes (AMOLED), microphone, speaker, calls, vibrators, etc. In one of the embodiments of the present invention, the user interface 306 may include, among other devices or elements, any device or all of the following devices: speaker, microphone, display, keyboard, touch screen, etc. In this regard, for example, the processor 302 may contain a diagram of the user interface, configured to control at least some functions of one or more user interface elements, such as speaker, ringer, microphone, display, and/or similar devices. The processor 302 and/or diagram of the user interface containing the processor 302 may control one or more functions of one or more elements of the user interface 306 by means of instructions of a computer program, for example, software and/or firmware stored in memory, for example, in at least one memory 304 and/or similar memory accessible by the processor 302.

In the exemplary embodiment of the present invention, the processor 302 may be constructed to contain the decoder 310 or manage. The decoder 310 may be any means such as a device or a circuit operating in accordance with software or implemented in APPA�atom the software or combination of hardware and software. For example, it may be a processor 302 that is running the software, the processor 302, made by ASIC or a FPGA configured to perform the operations described herein, or by their combination, in the configuration so the device or circuit to perform the corresponding functions of the decoder 310.

In an embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, detection of defects, eye color, such as red eyes, at least in one area of the eye that contains the eyes and the neighboring region. In an embodiment of the present invention, the area of the eye can be detected by detecting the first location or the area of the face, e.g. face 102 in image 100, and then determine the approximate location of the eyes, such as eyes 104 within the detected region 102 of the face. In an embodiment of the present invention, the face area may be determined using the technique of pattern recognition. The orientation of the face can also be obtained using the technique of pattern recognition.

In an embodiment of the present invention, the processor 302 is configured to provide �run device 300, together with the contents of the memory 304, and optionally other components described herein, to determine the approximate location of the eyes in the detected face area. The approximate area of the eye can be determined on the basis of coordinates of the face and the orientation of the face. In an embodiment of the present invention, the face area can be divided into four quadrants. For the front view portrait of a person one can assume that the upper left quadrant and upper right quadrant contain the eye, for example the left eye and right eye, respectively. For each possible eye treatment to compensate for the eye color defect may be performed independently. Possible eye may in the future be referred to as the first area of the eye. First the eye area may limit the first set of pixels. The first set of pixels may be used to determine pixels corresponding to the first region of the eye color defect.

In an embodiment of the present invention, the processor 302 is configured to perform the processing in the image space obtained from the bit array of RGB (red, green, blue). In another embodiment of the present invention, the processor 302 is configured to convert input image data available in other formats, in the RGB format for processing. For example,the input data for the image in the YUV format can be first converted into the RGB format, and then processed by the processor 302. In bit array RGB red region of space images can be highlighted, and the other region is suppressed. In an embodiment of the present invention, the specified image space can be obtained by calculating a differential image. In an embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, computing a first difference image. In an embodiment of the present invention, the first differential image is calculated on the basis of the difference between the luminance of the red pixel and the brightness of the green pixels from the set of pixels associated with the first region of the eye.

In this embodiment of the present invention, the calculation of the first difference image d1(x,y) can be represented by following mathematical expression:

,

otherwise d1(x,y)=0,

where r(x,y) - the brightness of a red pixel at point (x,y) on the image and g(x,y) is the luminance of the green pixel at point (x,y).

In another embodiment of the present invention, the first difference image d1(x,y) can be calculated as follows:

,

otherwise d1(x,y)=0.

The advantage of ICP�of litvania expressions (2) to calculate a differential image before the expression (1) is the expression (2) to compute d1(x,y) is more sensitive to lighter shades of red. Since red is in the area of the defect color of eyes or red eyes can have different colors, the use of multiple criteria (and related spaces) for their detection provides a very high percentage of recognition.

In another embodiment of the present invention, the first differential image is calculated based on the squared difference between the luminance of the red pixel and the maximum value of the luminance of the green pixel and blue pixel from the set of pixels associated with the first region of the eye in the image. In this embodiment of the present invention, the calculation of the first difference image d1(x,y) can be represented by the following mathematical expressions:

,

otherwise d1(x,y)=0

or

,

otherwise d1(x,y)=0.

The advantage of using the squared difference between the luminance of the red pixel and the maximum value of the luminance of the green pixel and blue pixel is that for those areas of the image where the difference between the luminance of the red pixel and the brightness of the green pixels high, as when the eye color defect, this brightness difference image may have a smaller value in the area of the skin.

� embodiment of the present invention, the first difference image d1(x,y), calculated in a specified way, can be normalized so that its values lie in the range from 0 to 255.

In an embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient. In an embodiment of the present invention a one-dimensional gradient of the first order in the x direction is calculated as follows:

Gx(x,y)=d1(x,y)-d1(x-1,y), if r(x,y)>g(x,y)

Gx(x,y)=0 if r(x,y)<=g(x,y).

At least one vertical projection of at least one of a gradient, for example the gradient Gx, can be calculated as follows:

Vp(x)=∑abs(Gx(x,y)), where the summation is produced for all values of y: 1<=y<=height.

Similarly calculates the one dimensional gradient Gy(x,y) of the first order in the y axis direction and the horizontal projection of Hp(y) associated with the specified gradient.

Although the above embodiment of the present invention was explained by using the gradients of the first order, the specified embodiment of the present invention can be explained or carried out using gradients of the second p�row, dispersions and any other classifiers that can be used to measure the difference/deviation of the values in the gradient image.

Based on the computed at least one projection processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, identifying at least one Central point. In an embodiment of the present invention, at least one Central point can be obtained from at least one projection. In an embodiment of the present invention, at least one projection may include at least one peak, wherein the location of the pixels corresponding to this at least one significant peak, is the Central point of the area of the eye color defect.

In an embodiment of the present invention, each of the horizontal projection and vertical projection can include a single significant peak, as shown in Fig.4A and 4B, respectively. In particular, Fig.4A and 4B illustrates a horizontal projection and vertical projection associated with at least one gradient according to the embodiment of the present invention.

In yet another embodiment of the present invention, at least one projection can be �one more significant peak. For example, in Fig.5A and 5B shows a horizontal projection and vertical projection associated with at least one gradient according to another embodiment of the present invention, wherein the horizontal projection contains two significant peaks, while the vertical projection contains one significant peak. In this case, there may exist two possible location of the eyes, i.e. (x1, y1) and (x1, y2). To determine the Central point in this case, you can consider a small region around the two points (x1, y1) and (x1, y2) and determine the maximum value for the difference image d1(x,y) in this field. Point having a larger maximum value d1(x,y), can be considered the Central point of the possible areas of the eye color defect.

The processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, computing the first area of the eye color defect using a center point and a plurality of red pixels of the first set of pixels. In an embodiment of the present invention to calculate a first area of the eye color defect can be used a method of increasing the area. For example, in one particular method, the capacity region can be considered those pixels of the first difference image d1(x,y), cat�which the brightness of pixels of the difference image is larger than a predetermined brightness threshold. Such pixels may be connected to the first Central point of the red eye area based on the connectivity of adjacent parts to form a related field with red eyes. In an embodiment of the present invention, these pixels can be associated on the basis of connectivity with one of the eight neighbors. In an alternative embodiment of the present invention, these pixels may be connected by connection with one of the four neighbors. The bound region as calculated in this manner can be viewed as an "instance" field defect eye color and check for signs of eye color defect.

In an embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, one or more verifications of the presence of the first eye color defect in the first area of the eye color defect. In yet another embodiment of the present invention the specified one or more verifications may be based on brightness of pixels in the first area of the eye color defect. In yet another embodiment of the present invention the specified one or more verifications can be carried out using the Naim�nichego bounding box thus the calculated area of the eye color defect may be enclosed in the smallest bounding rectangle. The neighboring region of the smallest bounding rectangle can be subdivided into a number of regions (N), and after that can be determined is whether the smallest bounding rectangle on the boundary of the first region of the eye color defect. If it is determined that the smallest bounding rectangle is located on the border of the selected area of the eye color defect, you can take the next passage to detect the presence of the eye color defect at the computed area of the eye color defect. However, if it is determined that the smallest bounding rectangle is not located on the border of the red eye area, calculates an average brightness of pixels in each of the neighboring areas of the smallest bounding rectangle. In addition, calculate the number (Nr1) neighboring areas, having an average brightness of pixels in the difference image d1(x,y) is less than a predetermined portion of the average brightness of the pixels enclosed in the first area of the eye color defect, and then compare Nr1 with a predetermined threshold number of neighboring areas. If it is determined that Nr1 is greater than a predetermined threshold number of neighboring areas, then we can determine the presence of the eye color defect at the computed first about�Asti eye color defect and subsequent verification of the computed first eye color.

In yet another embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, one or more verifications of the area of the eye color defect on the basis of its form.

In yet another embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, to identify an indicator of the reliability to indicate the presence of the first eye color defect at the computed first region of the eye color defect. In yet another embodiment of the present invention, the indicator reliability can be defined as the ratio of Nr1/N. If found, the indicator of confidence exceeds the first predetermined threshold, it indicates the presence of the first eye color defect in the first area of the eye color defect. In an embodiment of the present invention, the first predetermined threshold is 0.75. However, if it is determined that the indicator of reliability is less than the second preset threshold, it indicates the absence of the first eye color defect in the first area of the eye color defect. In an embodiment of the present invention, the second predetermined threshold is equal to 0.33.

In�the approach of implementation of the present invention in determining what is the value of the indicator of authenticity is located between the first predetermined threshold and second predetermined threshold, you can perform another pass. In yet another embodiment of the present invention, the second pass may be performed on the second set of pixels associated with the second region of the eye. The second area of the eye can be obtained by extending the boundaries of the first region of the eye. In yet another embodiment of the present invention in the second pass the second difference image d1(x,y) can be computed based on the squared difference between the luminance of the red pixel and the maximum value of the luminance of the green pixel and the blue pixel in the second region of the eye, is related to the brightness of a red pixel, and the brightness level of the second set of pixels. In another embodiment of the present invention, the second difference image d1(x,y) can be calculated based on the ratio of the square of the difference between the luminance of the red pixel and the maximum value of the luminance of the green pixel and the blue pixel in the second region of the eye to brightness of a red pixel, and the brightness level of the second set of pixels. In yet another embodiment of the present invention for calculating a second difference image d2(x,y) you can use the weight function p(r(x,y)), also known as correction functions�Oia. The correction function can be selected so that it takes small values for pixels with low brightness and large for pixels with high brightness, which reduces the probability of false detection in areas of low brightness.

During the second pass of the high value of the indicator of reliability indicates that the second area of the eye color defect calculated in the second pass, contains the eye color defect. However, if the indicator reliability is low, the second area of the eye color defect may not meet the eye color defect. In such a scenario, if it is determined that the projection contains two peaks, a new area of the eye color defect can be grown using a different center point corresponding to the peak, and verify the presence of the eye color defect.

In yet another embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, correction of the first eye color defect.

In yet another embodiment of the present invention, the specified image is an image of low resolution (LP). However, in some applications, the image resolution can be high, and accordingly, the detection of the eye color defect and to�breccia in the high-resolution image is a relatively complex and time consuming process. In this scenario, the detection of the eye color defect may be formed in the low-resolution image and its correction in high - resolution image.

Fig.6 shows a diagram illustrating the defects of the eye color in the picture is low resolution (LP) and the corresponding eye color defect in the high-resolution image (HP) according to the exemplary embodiment of the present invention. Typically, the capture device image, such as a camera, it is often desirable that the correction was made in the image high resolution image low resolution image high resolution retain in memory, and the low-resolution image display on the viewfinder.

Fig.6 illustrates the eye color defect in the image 610 low-resolution and matching the eye color defect in the image 650 high resolution. The image 610 is a low resolution contains the region of the eye color defect (shown by slanted hatching). In an embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, determining a first rectangle 612, bounding the first region of the eye color defect, and the second rectangle 614 written into the first area d�reaction eye color, in the image 610 low resolution. The second rectangle 614 represents the inner part of the area of the eye color defect. The area between the first rectangle 612 and the second rectangle 614 can be divided into many areas, such as the four quadrants. In yet another embodiment of the present invention in each of the four quadrants may be some defect with the eye color, such as part 616, and a part without a defect of the eye, such as part 618. So many parts, corresponding to these four quadrants can be generally attributed to the first set of peripheral areas.

In yet another embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, to determine the information about the eye color defect for low resolution images. In an embodiment of the present invention, the information about the eye color defect includes a set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image. In an embodiment of the present invention, the value of the set of identifiers includes the average brightness value of at least one part with defect C�ETA eye and at least one part without the eye color defect. Accordingly, in this embodiment of the present invention, the set of identifiers may include eight brightness values, with four brightness values correspond to the four parts of the eye color defect (such as part 616), and four values of brightness, corresponding to the four parts without defect eye color (such as 618 part) in the first area of the eye color defect.

In yet another embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, converting low resolution images into a high-resolution image. In yet another embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, the conversion information about the eye color defect, calculated on the basis of low resolution images, so as to make it suitable for use on high-resolution image. In yet another embodiment of the present invention during the conversion of the first rectangle 612 and the second rectangle 614 associated with the image 610 is a low-resolution, can be extended to the first rectangular�ka 652 and the second rectangle 654, associated with image 650 high resolution, as shown in Fig.6. The second rectangle 654 represents the inner part of the area of the eye color defect in the high-resolution image. However, after you convert the first rectangular 652 and the second rectangle 654 associated with the image 650 high resolution due to scaling may not touch the region of the eye color defect as in the low-resolution image 610. Accordingly, in the high-resolution image of the first rectangle 652 may be much larger than the area of the eye color defect, and the second rectangle 654 may be much smaller than the area of the eye color defect. In addition, parts of the eye color defect, such as a portion 616, and a part without a defect, eye color, such as part 618, the image 610 is a low-resolution, can be expanded in part with the eye color defect, for example 656, and part 658 without color defect of the eye, the image 650 high resolution.

In yet another embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, the image processing 650 high resolution. Image 650 high resolution can be processed by comparing the brightness of pixels raznostnoj� image to a plurality of second peripheral portion with the average values of the luminance difference image, calculated image 610 low resolution. In yet another embodiment of the present invention, the second peripheral portion of the plurality of second peripheral portion can be determined to contain a defect the eye color when the brightness of pixels in the differential image of the second peripheral portion in the image 650 high resolution closer to the average brightness value of the differential image for the corresponding part with the eye color defect than the average value of the brightness difference image of a part without a defect of the eye in the image 610 low resolution.

In yet another embodiment of the present invention, the processor 302 is configured to secure the device 300, together with the contents of the memory 304, and optionally other components described herein, correction of a defect of eye color.

Fig.7 shows the sequence of operations illustrating a method 700 of compensating eye color defects in an image, such as image 100, in accordance with a variant implementation of the present invention. The method is illustrated by the specified sequence of operations can be performed, for example, the device 300 shown in Fig.3. In an embodiment of the present invention, the image 100 may be an image of low resolution.

In block 702 computations�represent the first difference image. In this embodiment of the present invention, the first differential image is calculated on the basis of the difference between the luminance of the red pixel and the brightness of the green pixels from the set of pixels. In another embodiment of the present invention, the first differential image is calculated based on the squared difference between the luminance of the red pixel and the maximum value of the luminance of the green pixel and blue pixel from the set of pixels. The set of pixels is a pixel associated with the first region of the eye in an image of low resolution. The first difference image may indicate the region of the eye color defect in the first area of the eye.

In block 704, the first differential image is processed to calculate at least one gradient and at least one projection associated with the specified at least one gradient. In yet another embodiment of the present invention the specified at least one projection includes at least one horizontal projection and at least one vertical projection associated with at least one gradient. In yet another embodiment of the present invention, at least one gradient may be a one-dimensional gradient of the first order. Alternatively, said at least one gradient may include �radianti second order, dispersion, etc.

In block 706, based on the at least one projection define at least one Central point. In an embodiment of the present invention can be used to calculate the gradient of the first order in the x direction and the gradient of the first order in the direction of the y axis, then it is possible to calculate the vertical projection and horizontal projection, associated with these gradients. In yet another embodiment of the present invention the vertical projection and the horizontal projection may include at least one significant peak, with the location of the pixel that corresponds to at least one significant peak represents at least one Central point of area of the eye color defect on the image. In yet another embodiment of the present invention the specified at least one projection can have only one significant peak, as shown in Fig.4A and 4B. However, in some cases, the specified at least one projection may have multiple peaks, for example, as shown in Fig.5A. In this situation there may be two possible location of the eye, (x1, y1) and (x1, y2). To determine the Central point in this case can be considered a small area around the two points (x1, y1) and (x1, y2) and determine the maximum value for the difference image d1(x,y) in e�Oh area. Point having a larger maximum value d1(x,y) can be chosen as the Central point possible in the specified area of the eye color defect.

After determining the center point to possible areas of the eye color defect at block 708 calculates the first area of the eye color defect on the basis of the calculated center point and a plurality of red pixels of the first set of pixels. In yet another embodiment of the present invention, the region of the eye color defect can be calculated using the method of increasing the field.

In yet another embodiment of the present invention can determine the indicator of reliability. Indicator reliability can indicate the presence of the first eye color defect at the computed first region of the eye color defect. In yet another embodiment of the present invention, when it is determined that the value of the indicator of confidence exceeds the first predetermined threshold, it indicates the presence of the first eye color defect in the first area of the eye color defect. In another embodiment of the present invention, when the indicator value reliability less than the second preset threshold, it indicates the absence of the first eye color defect in the first area of the eye color defect. In yet another embodiment of the present invention, if the value indica�ora authenticity is located between the first predetermined threshold and second predetermined threshold, can be performed the next passage.

In an embodiment of the present invention, after determining the first region of the eye color defect, you can perform one or more verifications of the presence of the first eye color defect in the first area of the eye color defect. In yet another embodiment of the present invention the specified one or more verifications can be performed based on the method of the smallest bounding rectangle. In an alternative embodiment of the present invention the specified one or more verifications can be performed based on the shape of the region of the eye color defect.

In yet another embodiment of the present invention are used to correct the eye color defect detected in the first area of the eye color defect on the image. In yet another embodiment of the present invention, the eye color defect can be corrected by replacing the red pixels associated with the eye color defect at pixels natural eye colors such as natural black color pixels.

In yet another embodiment of the present invention, information about the eye color defect in the low-resolution image can be converted into a high-resolution image, and the eye color defect detected in the low resolution image, can be adjusted in every�the reflection high resolution. Method of compensation of the eye color defect by converting low resolution images into a high-resolution image is explained with reference to Fig.8.

Fig.8 shows the sequence of operations illustrating a method 800 compensating eye color defects in the images according to another embodiment of the present invention. In block 802 determines the region of the eye color defect at the low-resolution image (method of detecting defect eye color on the low-resolution image is described with reference to Fig.7). In yet another embodiment of the present invention, the low-resolution image can be formed by the high-resolution image.

In block 804 to determine the information about the eye color defect for low resolution images. In yet another embodiment of the present invention, information about the eye color defect includes a set of identifiers specific to the area of the eye color defect identified in the image with low resolution (as shown in Fig.6). The set of identifiers can be associated with many of the first peripheral portion of the first region of the eye color defect at the low-resolution image. In yet another embodiment of the present invention, the set of identifiers may include average values of the brightness difference image�of moving at least one part with the eye color defect and at least one part without the eye color defect in the low-resolution image.

In block 806, the information about the eye color defect from the low resolution images can be converted into a high-resolution image. A high-resolution image may include a second area of the eye color defect corresponding to the first region of the eye color defect in the low-resolution image, and a lot of second peripheral portion of the second region of the eye color defect corresponding to the set of the first peripheral portion of the first region of the eye color defect.

In block 808, the high-resolution image is processed to identify the eye color defect in at least one of the plurality of second peripheral portion based on a specified set of identifiers. In yet another embodiment of the present invention, the processing of high-resolution images involves comparing the brightness of pixels in the difference image to a plurality of second peripheral portion with respective average brightness of difference image computed from the low resolution images. In yet another embodiment of the present invention, the second peripheral portion of the plurality of second peripheral portion can be defined as a defect in eye color if the brightness of pixels in the differential image of the second peripheral portion in the high-resolution image is closer to the average value of the bright�ti differential image with at least one corresponding part of the eye color defect, than the average value of the brightness difference image corresponding part without the eye color defect in the low-resolution image.

In an embodiment of the present invention, it is possible to correct a defect the eye color in the picture is high resolution. In yet another embodiment of the present invention, correction of the eye color defect in the high-resolution image comprises replacing pixels corresponding to the eye color defect at pixels matching the natural color of their eyes in the high-resolution image.

Without limiting the scope or application of the claims, the technical result of one or more embodiments of the present invention is to compensate for the defects of the eye color in the image. Another technical effect of one or more embodiments of the present invention lies in the detection of eye color defects on the low-resolution image and a subsequent correction of the detected defects the eye color on the image is low resolution. Another technical result of one or more embodiments of the present invention lies in the detection of eye color defects on the low-resolution image and a subsequent correction of the detected defect color image of high resolution. Another t�detailed technical result of one or more embodiments of the present invention is very accurate in compensating eye color defects. In addition, since the present invention uses a property of redness of the eye color defect, the difficulty in defining the eye color defect is significantly reduced. In addition, the use of multiple passes for different criteria ensures reliable detection of defects of the eye colors of all shades and hues, providing a very high detection rate. In addition, the calculation of the degree of reliability with each pass allows early exit from the process if at any stage it becomes clear that the calculated area of the eye color defect may be a defect of eye colors with a high degree of reliability, which reduces the amount of computation.

Available methods and devices, allowing you to make a choice between many possible areas of the eye where the above-mentioned projection have many significant peaks. In addition, for cases where you want to remove the defect the eye color on the image is very high resolution, a method of detection of the defect on the low-resolution image and use this result for the effective and accurate correction of this defect on the high-resolution image.

Various embodiments of the present invention described above may be implemented as software, hardware, log�key applications, or a combination of software, hardware and application logic. Software, logic, application and/or hardware may reside in at least one memory, the at least one processor, device, or computer software product. If necessary, the portion of the software, application logic and/or hardware may reside on the device that detects a defect of the eye color, or part of the software, application logic and/or hardware may reside on the device, correcting the eye color defect. In one embodiment of the present invention, logic applications, software or a set of instructions stored on any of a variety of known computer readable media. In the context of the present description of computer-readable media may be any media or means that can contain, store, communicate, propagate or transport the instructions for use system, device or means for executing instructions, such as a computer, an example of which is described and illustrated herein, or for use in connection with the listed devices. Machine-readable medium may include machine-readable media can be any media interedsted, that can contain or store the instructions for use system, device or means for executing instructions, such as a computer, or for use in connection with the listed devices.

If necessary, various features discussed above can be performed in a different order and/or simultaneously with each other. In addition, if necessary, one or more of the above-described functions may be optional or may be combined.

Although various aspects of the invention set forth in the independent claims, other aspects of the invention comprise other combinations of features are described variants of the invention and/or the features specified in the dependent claims with the features specified in the independent claims, and not only those combinations explicitly set out in the claims.

In addition, it should be noted that although the above described embodiments of the present invention, this description should not be construed as limiting the invention. Can be made of changes and modifications within the spirit of the invention defined by the claims.

1. Method of detecting red-eye in an image, including:
calculating a first difference image based on the difference between the brightness of the red� pixels and brightness of the green pixels from the set of pixels, wherein said set of pixels associated with the first region image;
processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient;
determining at least one Central point based on the specified at least one projection and
calculating a first region of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

2. A method according to claim 1, wherein the specified image is an image of low resolution (LR).

3. A method according to claim 1, wherein the specified at least one projection includes at least one horizontal projection and at least one vertical projection associated with the specified at least one gradient.

4. A method according to claim 1, wherein the definition of the specified at least one Central point includes determining at least one significant peak at least one projection, wherein the location of the pixels corresponding to the specified at least one significant peak is at least one Central point of the first region of the eye color defect.

5. A method according to claim 1, also including you�olnine one or more verifications of the presence of the first eye color defect in the first area of the eye color defect, when the calculated first region of the eye color defect.

6. A method according to claim 5, also including the definition of the indicator of confidence in the indications of the presence of the first eye color defect in the first area of the eye color defect, in this case,
determine the presence of the first eye color defect in the first area of the eye color defect, if the indicator of reliability is larger than the first predetermined threshold, and
determine the absence of the first eye color defect in the first area of the eye color defect, if the indicator of reliability is less than the second predetermined threshold.

7. A method according to claim 6, also comprising the following passage, if it is determined that the indicator of reliability lies between the first predetermined threshold and second predetermined threshold.

8. A method according to claim 5 or 7, also including a correction of the first eye color defect.

9. A method according to claim 2, also including:
definition of the information about the eye color defect for low resolution images, wherein said defects of eye colors contains a set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image and the first region of the eye color defect contains the first eye color defect;
the transformation of information about the eye color defect, calculated on the basis of the image�ia low resolution so, to make it suitable for use on high-resolution image (HR), a high-resolution image contains
the second area of the eye color defect corresponding to the first region of the eye color defect in the low-resolution image, and
a lot of second peripheral portion of the second region of the eye color defect corresponding to the set of the first peripheral portion of the first region of the eye color defect, and
processing high-resolution images to identify the eye color defect in at least one of the plurality of second peripheral portion based on a specified set of identifiers.

10. Method of detecting red-eye in an image, including:
the definition of a first region of the eye color defect in the image is low-resolution (LR) image low resolution image formed from a high-resolution (HR);
definition of the information about the eye color defect for low resolution images, and the information about the eye color defect includes a set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image, wherein the value of the set of identifiers includes the average brightness value of at least one part with the eye color defect and at least one� parts without defect the eye color for the specified set of the first peripheral portion;
the transformation of information about the eye color defect, calculated on the basis of low resolution images, so as to make it suitable for use on high-resolution image, a high-resolution image contains
the second area of the eye color defect corresponding to the first region of the eye color defect in the low-resolution image, and
a lot of second peripheral portion of the second region of the eye color defect corresponding to the set of the first peripheral portion of the first region of the eye color defect, and
processing high-resolution images to identify the eye color defect in at least one of the plurality of second peripheral portion based on a specified set of identifiers, wherein the treatment is carried out by comparing the brightness of pixels of the difference image to a plurality of second peripheral portion with the average values of the brightness difference image computed from the low resolution images.

11. A method according to claim 10, also including the correction of the eye color defect identified in at least one of the plurality of second peripheral portion in the high-resolution image.

12. Device for detecting red eye in the image that contains:
at least one processor and
at least one memory, containing�second computer program code thus specified at least one memory and computer program code configured to, with the specified at least one processor to enforce the specified device at least the following:
computing a first difference image based on the difference between the luminance of the red pixel and the brightness of the green pixels from the set of pixels, wherein said set of pixels associated with the first region image;
processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient;
identifying at least one Central point based on the specified at least one projection and
computing the first area of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

13. The device according to claim 12, wherein said image is an image of low resolution (LR).

14. The device according to claim 12, in which the specified at least one projection includes at least one horizontal projection and at least one vertical projection associated with the specified at least one gradient.

15. Mouth�austo according to claim 12, which is also provided, at least partially, the execution of the specified device identifying at least one Central point by identifying at least one significant peak at least one projection, wherein the location of the pixels corresponding to the specified at least one significant peak, represents the specified at least one Central point of the first region of the eye color defect.

16. The device according to claim 12, which is also provided, at least partially, the execution of the specified device one or more verifications of the presence of the first eye color defect in the first area of the eye color defect when the calculated first region of the eye color defect.

17. The device according to claim 16, which is provided, at least partially, the execution of the specified device to determine an indicator of the reliability to indicate the presence of the first eye color defect in the first area of the eye color defect, in this case,
determine the presence of the first eye color defect in the first area of the eye color defect, if the indicator of reliability is larger than the first predetermined threshold, and
determine the absence of the first eye color defect in the first area of the eye color defect, if the indicator of reliability is less than the second predetermined threshold.

18. A device for detecting�of fishing red eye in the image, contains:
at least one processor and
at least one memory containing computer program code, wherein the specified at least one memory and computer program code configured to, with the specified at least one processor to enforce the specified device at least the following:
determine the first region of the eye color defect in the image is low-resolution (LR) image low resolution image formed from a high-resolution (HR);
determining information about the eye color defect for low resolution images, and the information about the eye color defect includes a set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image, wherein the value of the set of identifiers includes the average brightness value of at least one part with the eye color defect and at least one part without defect the eye color for the specified set of the first peripheral portion;
convert information about the eye color defect, calculated on the basis of low resolution images, so as to make it suitable for use on high-resolution image, a high-resolution image contains
W�the ROI region of the eye color defect, corresponding to the first region of the eye color defect in the low-resolution image, and
a lot of second peripheral portion of the second region of the eye color defect corresponding to the set of the first peripheral portion of the first region of the eye color defect, and
image processing high resolution for the detection of the eye color defect in at least one of the plurality of second peripheral portion based on a specified set of identifiers, wherein the treatment is carried out by comparing the brightness of pixels of the difference image to a plurality of second peripheral portion with the average values of the brightness difference image computed from the low resolution images.

19. Machine-readable media containing a set of instructions that when performed by one or more processors provide performance of the device at least the following:
computing a first difference image based on the difference between the luminance of the red pixel and the brightness of the green pixels from the set of pixels, wherein said set of pixels associated with the first region image;
processing the first difference image for computing at least one gradient and at least one projection associated with the specified at least one gradient;
definition�Oia at least one Central point based on the specified at least one projection and
computing the first area of the eye color defect on the basis of said at least one Central point and a plurality of red pixels from the specified set of pixels.

20. Machine-readable medium according to claim 19, in which the specified image is an image of low resolution (LR).

21. Machine-readable medium according to claim 19, in which the specified at least one projection includes at least one horizontal projection and at least one vertical projection associated with the specified at least one gradient.

22. Machine-readable medium according to claim 19, which also enforced the specified device at least partially defining at least one Central point by identifying at least one significant peak at least one projection, wherein the location of the pixels corresponding to the specified at least one significant peak, represents the specified at least one Central point of the first region of the eye color defect.

23. Machine-readable medium according to claim 19, which also enforced the specified device, at least partially, one or more verifications of the presence of the first eye color defect in the first area of the eye color defect when the calculated first region of the eye color defect.

p> 24. Machine-readable medium according to claim 23, which also enforced the specified device, at least partially, determine an indicator of the reliability to indicate the presence of the first eye color defect in the first area of the eye color defect, in this case,
determine the presence of the first eye color defect in the first area of the eye color defect, if the indicator of reliability is larger than the first predetermined threshold, and
determine the absence of the first eye color defect in the first area of the eye color defect, if the indicator of reliability is less than the second predetermined threshold.

25. Machine-readable media containing a set of instructions that when performed by one or more processors provide performance of the device at least the following:
determine the first region of the eye color defect in the image is low-resolution (LR) image low resolution image formed from a high-resolution (HR);
determining information about the eye color defect for low resolution images, and the information about the eye color defect includes a set of identifiers associated with many of the first peripheral portion of the first region of the eye color defect in the low-resolution image, wherein the value of the set of identifiers includes the average values of brightness on IU�Isha least one part with the eye color defect and at least one part without defect the eye color for the specified set of the first peripheral portion;
convert information about the eye color defect, calculated on the basis of low resolution images, so as to make it suitable for use on high-resolution image, a high-resolution image contains
the second area of the eye color defect corresponding to the first region of the eye color defect in the low-resolution image, and
a lot of second peripheral portion of the second region of the eye color defect corresponding to the set of the first peripheral portion of the first region of the eye color defect, and
image processing high resolution for the detection of the eye color defect in at least one of the plurality of second peripheral portion based on a specified set of identifiers, wherein the treatment is carried out by comparing the brightness of pixels of the difference image to a plurality of second peripheral portion with the average values of the brightness difference image computed from the low resolution images.



 

Same patents:

FIELD: physics, video.

SUBSTANCE: invention relates to apparatus for selecting highly detailed objects. The apparatus comprises an input realisation storage unit, a control unit, a Canny edge detector unit, a unit for determining a common detailing coefficient, a sliding window selection unit, a unit for determining the detailing coefficient in the sliding window, a comparison unit, an output realisation storage unit and a clock pulse generator.

EFFECT: selecting highly detailed objects on images and video sequences on an arbitrary low-detailed background.

1 dwg

FIELD: physics, computer engineering.

SUBSTANCE: group of inventions relates to means of orientation in real space. The device comprises a device for capturing an image of a predetermined area, a unit for transforming a view point, which creates a bird's eye view image, a computer for calculating luminance difference between two pixels near positions along a vertical imaginary line, which stretches in a vertical direction in real space, a three-dimensional object detector which detects said object based on contiguities of luminance differences of corresponding calculated positions.

EFFECT: high accuracy of detecting a three-dimensional object in surrounding space.

10 cl, 25 dwg

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to digital imaging means. The system comprises a pixel pre-processing unit comprising logic for collecting automatic white balance statistics and automatic exposure setup statistics, statistics collection logic comprising an input, colour space transformation logic and a set of pixel filters. The system also comprises control logic configured to determine a set of point white balance gain coefficients to apply to multi-colour raw frame pixels based on accumulated colour sum values.

EFFECT: high quality of the resultant image by eliminating artefacts in a full-colour image.

20 cl, 74 dwg

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to means of processing digital images. In the method, an image in the RGB colour space is converted to a grayscale; processed with Laws energy characteristics by folding the matrix of the source image with nuclei of the type of detecting an image in the form of ripples, detection of spots, detection of waves and detection of edges; for each of the obtained image, spots, corners and line edges are detected by a SURF method to find a local point using Hessian matrix.

EFFECT: detection of local features on an image.

2 cl, 2 dwg

FIELD: physics.

SUBSTANCE: method includes forming and storing reference and current contour images, rendering said images in the form of half-tone distant images, where pixel brightness is proportional to the distance to the nearest contour, selecting areas possibly containing an object on the images by scanning with a contour reference while counting for each reference position the brightness sum of the points on the distant half-tone image covered by the contour reference, making points where the brightness sum is less than a given threshold, at the selected areas of the images, calculating a cross-correlation function of current and reference images, comparing the selected areas of the current images with the reference half-tone distant images of said areas, determining the reference position on the selected areas where the extremum of the cross-correlation function is reached, determining the position of the object from the position of the extremum of the cross-correlation function.

EFFECT: high accuracy of detecting an object on an image.

13 dwg

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to image content processing means. The method involves browsing the object to evaluated and identifying features of an undesirable image of a first group - dynamic features and of a second group - static features. If a feature of a first group or a second group is identified, features of a third group (adverse health consequences) and of a fourth group (shooting conditions) are identified. The method involves formalised analysis of the evaluation of the object based on said groups of features, determining a quantitative expression of the identified features based on factorial analysis taking into account the factorial load coefficient, summation of quantitative indicators of the identified groups of features and determining presence of an undesirable image.

EFFECT: high efficiency of evaluating image content.

4 cl

FIELD: physics.

SUBSTANCE: in the method, by selecting on an original infrared image reference points representing values of ice thickness and brightness thereof; using values of brightness and the calculated functional relationship between a virtual relief and the relief of the brightness field on the infrared image obtained from an artificial earth satellite; for the original infrared image, brightness of "warm" and "cold" points is calculated; the analysed image is displayed in form of a spatial distribution of matrix elements corresponding to brightness values of each image pixel; ordered structures of the brightness field are selected and then displayed in form of a colour painting.

EFFECT: enabling monitoring of the state of ice cover on water areas by determining thickness of the ice cover, averaged on a local resolution element.

3 cl, 2 dwg

FIELD: information technologies.

SUBSTANCE: system comprises a module of detection of areas of interest, which analyses image data and determines the position of areas of interest, a module of detection of criteria of areas of interest, which calculates criteria of areas of interest; at the same time a list of areas of interest arrives to the module inlet, a module of generation of visual objects, which creates visual objects of effect, a module of detection of criteria of ambient sound, which calculates parameters of ambient sound, a module of generation of animated frames, which generates frames of effect animation, combining the initial static image and visual objects, modified on the basis of current criteria of ambient sound, a display device, providing animation frames to a user, which are produced from the module of generation of animated frames.

EFFECT: exclusion of repetitions of generated frames of animation during reproduction and provision of compliance of frames with background ambient sound.

9 cl, 7 dwg

FIELD: medicine.

SUBSTANCE: system comprises a digital image input unit, a system parameter input unit, a final image reader and memory; an image morphological filtration module, a separator for local minima of image intensity; a measurer of distance from each image pixel to the nearest background pixel; a separator for local maxima of image intensity; an OR element; a noise object border filter; a morphological image gradient detector; a border mask unit of the required objects by segmentation; a calculator of selected object coordinates; a unit of obtaining the optical and geometrical characteristics of the detected objects and the coordinate database, the optical and geometrical characteristics of the detected objects.

EFFECT: higher accuracy of object selection on the images containing a number of irregularly distributed small spatial objects.

5 cl, 8 dwg

FIELD: information technology.

SUBSTANCE: in the method, an extend function is applied to image data to obtain extended data having a wider dynamic range than the original data. An expansion card is also obtained, which contains data on brightness of regions associated with pixels on the image. The method then merges the original image data with the extended data in accordance with the expansion card, yielding enhanced image data.

EFFECT: extension of image dynamic range.

25 cl, 3 dwg

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to image forming apparatus with an identity authentication function. Disclosed is an image forming apparatus which includes an image forming unit. The apparatus also includes a recording unit, configured to record a plurality of data on characteristic features, up to the maximum limit, in non-volatile memory with reference to the same object. The apparatus also includes an authentication unit, configured to identify whether an object detected based on an image captured by the image forming unit is a recorded object based on data on characteristic features recorded in the nonvolatile memory. Furthermore, the apparatus includes a selection unit, configured to select one of the objects recorded in the nonvolatile memory, for which new data on characteristic features were recorded through the recording unit.

EFFECT: high accuracy of authenticating an object by selecting new data on characteristic features.

18 cl, 22 dwg

FIELD: physics.

SUBSTANCE: disclosed is an apparatus for determining outlines of objects on an image. The apparatus comprises an image element sensor and the following units for: generating threshold values of speed and acceleration of change in intensity of the image elements, creating an eight-connected neighbourhood of the image elements, successive selection of triads in the eight-connected neighbourhood of an image element, memory for said triads, differentiating triads and a decision unit. The unit for generating threshold values of speed and acceleration of change in intensity of the image elements is connected to the output of the image element sensor. The unit for creating an eight-connected neighbourhood of the image elements is connected to the output of the unit for generating image elements and is connected to series-connected unit for successive selection of triads in the eight-connected neighbourhood of an image element, memory unit for triads of the eight-connected neighbourhood of an image element, triad differentiation unit and decision unit.

EFFECT: high noise-immunity of recognising an image outline by estimating the speed and acceleration of change of colour intensity in the eight-connected neighbourhood of an image element.

7 dwg

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to digital image processing means. In the method, said operator is formed from linear structure-forming elements with different orientation parameters relative the raster of an image of equal length; each filtered image is obtained through interaction of the linear structure-forming element of the composite morphologic operator with the original image; pixel brightness in the filtered image is obtained by performing, for each pixel of the original image, three morphologic operations of interaction of the original image with the linear structure-forming element.

EFFECT: high accuracy of highlighting boundaries of complex-structure images owing to formation of multiple direction-filtered images from an original half-tone image via local processing with a composite morphologic operator.

6 dwg

FIELD: physics.

SUBSTANCE: system (200) has breaking unit (210) for breaking an image into a plurality of image areas, each image area displaying an area of the anatomical structure, and an adaptation unit (220) for adaptation of the referenced image to the evaluation function-based image, wherein the evaluation function is a function of parameters of the reference image and index numbers of the image and relative positions thereof on the image, and based on a criterion which must be satisfied by the index numbers of the evaluation function, wherein the evaluation function is determined based a plurality of image areas. Determination of the evaluation function based on a plurality of image areas enables to calculate optimum values of input into the value of the evaluation function within each image area.

EFFECT: high reliability of automatic detection of a cardiac muscle in short axis layer images.

11 cl, 9 dwg

FIELD: information technology.

SUBSTANCE: during training, a random number N with length of n bits is obtained; n artificial neurons are created; neuron inputs are randomly connected to memory cells with data on inclination of dermal ridges of fragments of examples of the classified fingerprint picture; further, the neurons are trained to output a random number N when presenting examples of the classified picture; the random number N is the hashed and the reference hash value is stored, as well as the connection of the neural network and its training parameters; and during authentication, there is repeated breaking up of the scanned picture into fragments, calculation of the average inclination on the fragment and the calculated value is transmitted to inputs of the neural network; the hash result is calculated at the output of the neural network and then compared with the reference value; in case of a match, personal authentication is continued, otherwise the authentication procedure is stopped once the time allocated for an authentication session runs out.

EFFECT: safeguarding the finger print picture and its biometric parameters which are used during biometric authentication, and providing conditions wherein it is impossible to find the owner of biometric parameters from biometric authentication data, while also cutting the average time of loading the authentication server.

2 cl, 3 dwg

FIELD: information technology.

SUBSTANCE: fragments of a digital image similar top human faces are detected in all regions of the digital image; the indicator of the fraction of pixels of skin colours inside the region of a fragment is determined on said fragments; a face is detected if a given threshold value is exceeded; if the threshold value of the fraction of skin colour inside the region of the found face is exceeded, the initial colour palette of human skin is adjusted; outlines of the object of the colour skin are determined; an indicator of geometric similarity of the found outline with typical body outlines is calculated; fragments similar to typical naked body parts are searched; the image is analysed for pornographic visual content.

EFFECT: high accuracy of analysing content of a digital image.

16 cl, 2 dwg

FIELD: information technologies.

SUBSTANCE: method is proposed to identify contours of moving objects, including detection of moving pixels according to a principle of inter-frame subtraction, detection of direction of movement of detected pixels with account of direction of movement of adjacent pixels, generation of moving objects by combination of adjacent pixels with one direction and availability of pixels in an eight-connected vicinity, spatial differentiation of detected objects by a Sobel operator, skeletonisation of contour lines on a gradient image by the method of suppression of points of non-maximum brightness, threshold processing of a skeleton gradient image on the basis of maximum and minimum of enthropy.

EFFECT: higher accuracy in detection of areas of moving objects and higher speed for identification of image contours.

1 dwg

FIELD: physics; computer engineering.

SUBSTANCE: invention relates to computer engineering and can be used for detecting image contours and subsequent input of the result into a computer. The device has a bus controller, RAM for reading (recording) intermediate image processing results, image filter unit, a contour detecting unit and a microprocessor.

EFFECT: more accurate detection of image contours.

1 dwg

FIELD: physics.

SUBSTANCE: invention relates to location of part of a document captured in an image. A method is proposed for determining part of a document, which corresponds to the captured image, when the user makes a flourish in the document. Location of some of the images will be determined, for example by analysing pattern on the capture image, or from results of pixel-by-pixel comparison of the image with the document. Location of other images will be determined through segmentation of the sequence of images in a group of corresponding forms of the flourish. Information relating to localised images in the segment can then be used for determining position of non-localised images in the segment. The search area of a non-localised image in the document is established based on the position of the previous localised image and maximum speed or maximum acceleration of the pen. Turning and scale of the non-localised image are evaluated, as well as turning and scale of the localised image, and the non-localised image is deformed using this turning and scale. Pixel-by-pixel comparison between the deformed non-localised image and the search area in the document is done. If deformation of the non-localised image is successfully matched, parametres of image transformation are further specified.

EFFECT: increased efficiency of locating part of a document.

10 cl, 25 dwg

FIELD: physics, image processing.

SUBSTANCE: invention is related to digital processing of image in process of scanning and copying, and especially to the field of colour and black-and-white text segmentation, when text is automatically extracted from scanned document. According to the present invention, method of text segmentation by colour criterion consists in performance of the following operations: initial image is broken down into non-overlapping units of pixels; new image Z is generated, in which each pixel represents corresponding unit of initial image; in process of scanning serial units are selected from initial image; classification is carried out for current unit by criterion "monochromatic/colour" in space of opposite colours; initial colour unit RGB is broken down into monochromatic units R, G and B; detector of Laplacian-Gaussian edges is applied with specified threshold T to monochromatic unit; number of edges is calculated for each pixel in Z; classification of "text/non-text unit" is performed by comparison of edge number with set threshold C; classified channels are combined, using logical operator OR.

EFFECT: provision of universal approach to preliminary processing of initial document, providing for faster printing of copy and saving of multifunctional device resources.

4 cl, 3 dwg

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to computer engineering and can be used in image analysis and processing systems and in digital television. The device comprises a preprocessing unit 1, a control unit 2, a first convolution computing unit 3, a second convolution computing unit 4, a second Gabor mask generating unit 5, a first Gabor mask generating unit 6, a first binarisation unit 7, a second binarisation unit 8, a first mask storage unit 9, a second mask storage unit 10, an adder 11, a post-processing unit 12, a resultant defect mask storage unit 13.

EFFECT: detecting the position of defects on archival photographs with insufficient prior information on statistical characteristics of distortions.

2 dwg

Up!