Method (variants) and device for identification of digital video data from several sources, television surveillance system

FIELD: analysis of television images, possible use in video surveillance systems.

SUBSTANCE: in accordance to the invention, a set of digital video data from first source, representing first image, is identified as standard video data of first source. Then a second set of video data is read, which represents current image. Difference ratio is computed using standard digital video data and current set of digital video data. If difference ratio exceeds a threshold, a query is shown to system user on the display to receive a response about whether current digital video data belong to identified source, or originate from new source. If response points at new source, then current set of digital video data is dispatched for storage into second memory cell, connected to second source. Then the current set of digital video data is identified as standard digital video data for second source.

EFFECT: division of digital video data outputted by several sources.

5 cl, 9 dwg

The technical field to which the invention relates.

The present invention relates to the analysis of television images. In particular, the present invention relates to the separation of non-indexed flow signal of the television image.

The level of technology

In the prior art technique known combining the output signals from several cameras with the formation of the video signal representing a composite signal composed of the output signals of the cameras. Usually the combining images is sequentially when one or more frames of any camera is inserted into the video (sequence), then the multiplexer adds one or more frames from the next camera. The number of frames taken from each camera is determined by the fed to the multiplexer signal synchronization, the number of cameras and shooting speed. Also in the prior art uses a digital representation of the video signal and the identification of the images received from each camera in the combined signal. In the prior art, however, this procedure is performed sequentially, that is, the images from one camera, identified by mapping across the array of digitized video. In other words, to obtain signals for all N cameras required N promo the ditch video. In addition, in such systems is only the direct comparison of the images. For example, after the first image are identified and linked (associated) with the first camera, the comparison is performed for each of the other images in the video signal. If during shooting will be changed in the field of view of the first camera and there will be significant differences in comparison with the first image, the system will reject the image as not identified, even if this image should be associated with this camera. Thus, in known systems are not reliable recognition of movements in front of the camera. Moreover, such systems are not suitable for cameras, leading panoramic shooting that scene in the field of view is constantly changing. Thus, it requires a system that makes it possible to conduct parallel processing of the signals from all the cameras on digitized video signal in a single pass. Also need a way designed to work with cameras panoramic shooting, and the corresponding system. In addition, you want a system in which a reset frame when splitting the signal and provides the detection of motion in a field of view of the camera.

Disclosure of inventions

In one embodiment of the invention proposes a method of sec is of the digital video data, coming from multiple sources, which combine images, no indexing or missing identification data that is used to separate the sources. Sources usually are cameras that can be fixed or movable - swing that is commonly used in surveillance systems. The first set of digital video data representing the first image is retrieved from a storage device or from videomagician account. The first set of digital video data remain in the memory cell connected (associated) with the first source. The first set of video data is also identified as an exemplary video data representing the first source. Then get a second set of digital video data representing the current image. Calculate the differential ratio using standard digital video data and a set of digital video data of the current image. If the differential ratio exceeds a threshold ("threshold"), the user of the system is presented on the display request to get an answer about whether the current digital video data to the identified source, or they come from a new source. In that case, if the query response indicates that the current set of digital video data belongs to a new source, those who use a set of digital video data save in the second memory cell, associated with the second source. Then, this current set of digital video data is identified as exemplary video data for the second source.

If the differential ratio is below the threshold, the current digital video data remain in the memory cell associated with the first source, the current set of digital video data is now becoming exemplary video data for the first source. In some embodiments, execution, before the calculation of the differential relations produce normalization of digital video data to compensate for differences in brightness. If you have multiple sources, and the sources are identified and are associated exemplary frames, differential ratio calculated for each reference frame, and each differential ratio is compared with a threshold. If the differential ratio, corresponding to an exemplary frame of the source is below the threshold, the current frame of the digital video data is the new model frame for this source.

In some embodiments, the digital video data is stored as a reference frame, is subjected to compression before storing. In such embodiments, the execution, the current digital video data representing the current video frame, are subjected to compression before the calculation of the differential relations.

In other embodiments, the implementation of the Oia, the user can send the signal, containing information representing a part of the image. Data in this part of the image represent the data that is used for comparison in which the same part of the current frame and of each of the reference frame used in the calculation of differential relations.

The above-described method is characterized by the fact that the identity of all the frames of the data array to the source can be determined in a single pass through the array of data. The described methodology can be implemented in computer software using computer code, computer readable medium.

This method can be carried out in the device, consisting of modules. The device may be part of a system that includes a camera and a processor for receiving each of the video signal, and combining the signal in the combined signal. The system may additionally include a memory for storing the United and divided video signals. In this system, the processor receives a request for separation processor separates the combined signal in a single pass on the combined signal. The system may include a receiving module for receiving a first set of digital video data representing the first image, and receive the current set of video data representing the current image. Further, the system should have the module PA is ATI, the computational module and the module requests. The processing module calculates a differential relation with the first set of digital video data and the current set of digital video data, and the memory module stores a first set of digital video data in a memory cell associated with the first source. Query engine generates a query on the display, requiring a response concerning the supplies digital video data identified the source or origin from a new source, if the differential ratio exceeds the threshold.

The memory module also sends the current set of digital video data for storage in the second memory cell associated with the second source, in accordance with the received response indicating that the current set of digital video data is associated with the new source. When the difference has a value below the threshold, the memory module sends the current digital video data for storage in a memory cell associated with the first source. The system may also include a normalization module for normalizing the digital video data to compensate for differences in brightness before displaying the digital video data in a computing module for calculating a differential relationship. In this system, the computational module is a means of calculating a differential relations for each of the reference set of digital to the s video representing the source, and the current set of digital video data, and the request is a means of comparing a differential relationship with the threshold value.

Brief description of drawings

The features of the invention will become clearer upon reading the following detailed description together with the attached drawings, on which:

Figure 1 represents the configuration of a system for implementing the first version of the runtime to separate television images;

Figure 2 is a block diagram illustrating the sequence of separation television frames from multiple cameras from a stream of frames of the television image;

Figure 3 is a block diagram illustrating additional features of the television division personnel;

Figure 4 represents the frame of the comparison, shown on the display, and the rectangle with lines shows the user-selected area;

Figure 5 represents the current frame shown on the display, and shows that the same user-selected area, as in figure 4, is used for comparison.

6 shows three cameras (camera 1, camera 2 and camera 3)generating a sequence of combined television frames where the image from each camera is introduced in the sequence of time intervals;

7 and 8 shows the selected user is the user area, of interest, in which reproduced the date and time;

Fig illustrates an embodiment of the invention, in which the modules are presented in the form of hardware, performing the functions of the proposed method.

The implementation of the invention

In the following text, the term "digital signal" means the representation in digital form sequences of images that can be consistently reproduced on the display. Typically the digital video signal includes multiple frames, each of which plays a separate image. Frames, in turn, can be divided, representing a set of pixels. In the following text, the term "pixel" refers to a single image point. The greater the number of pixels contained in the image, the higher the resolution of the television image. Resolution is usually described by the number of pixels in length and width, for example, resolution 800×600 corresponds to 800 pixels on the image's length and 600 pixels across the width of the image.

In an embodiment of the present invention, the television image in the stream of images generated by multiple sources of video are separated so that the image from a particular source may buttparade, analyzed and displayed for viewing.

Figure 1 shows the system configuration for implementing the option of carrying out the invention. Figure 1 shows several cameras (Camera a, Camera b and Camera). Each camera produces a video signal containing video information, which is combined and stored either in digital form in a memory device integrated with the processor 20, either in digital or analog form on the tape 25. Combined (multiplexed) information does not contain any header that identifies the source, nor any indexes to identify the source. Then the combined information is shared (demultiplexers) processor. If the video information is represented in analog form, it is first digitized by an analog-to-digital conversion, familiar to mid-level professionals. After digitizing a sequence of digital data is passed to the processor. The image signals may or may not be subjected to compression and presented in the format of fields or frames. Similarly, the digital signal can be represented by any of the available transfer methods colors, including RGB and YUV. The processor performs operations on the separation of the digital video signal in accordance with the flowchart in figure 2 and 3. It should be borne is mind, in the framework of the present disclosure, the term "frame" is used to denote the television image, which must be reproduced within a given period of time. The present disclosure is equally applicable to television fields that are generated for display, use for imaging through a lower-scan.

Figure 2 and 3 shows the separation of signals from multiple sources 100. The first frame of a television signal in a digital video signal is analyzed and stored in the memory cell 110. The default frame is denoted as a frame of comparison for the first source, which in this embodiment is a camera 1. The frame comparison represents the frame, which will be mapped to the current frame. In one embodiment of the frame, which can be a frame in the NTSC standard, before the comparison operation is subjected to reduction to, for example, 1/8 of its original size before storing it in memory. Reducing the size of the frame can be carried out simply with the exception of pixels. For example, to reduce the number of pixels in 4 times the image size 800×600 pixels should be excluded pixels in one, leaving 400×300 pixels. In another embodiment, the data volume can be reduced by averaging. For example, every eight R is Dov eight pixels can be averaged so, that 64 pixels is reduced to one pixel. Then, from the digital video signal going to the next frame (current frame).

Then for frame comparison to highlight the selected area of the user. This selected area is used to map to the selected area of the frame, comparing with the same selected area of the current frame. In some embodiments, executing the selected default scope is the entire frame, and no interaction with the user system is not required. Before comparing frame comparison and the current frame is the normalization of selected areas of frames to exclude from the comparison of any changes in the brightness/intensity. For example, the computation of the values of the average intensity of the pixels in the selected area of the user in the frame of the comparison and in the same area in the current frame. Then the intensity value for each pixel is divided by the average intensity value for the corresponding selected area of the user. This step is the normalization of values in connection with any change of brightness, for example, due to sudden flashes of light. For example, if the camera is monitoring the ATM and the ATM approaching car lights which suddenly illuminate the monitored space, such as the lighting changes will be taken into account. Like normal the grouting can be performed by any known method, including the use for the selected user area instead of the average intensity of the RMS value.

Then, using the selected area of the user, at step 130 produces comparing the current frame with the same area of the frame comparison. Figure 4 and 5 shows an example where 4 shows the frame 200 comparison displayed on the display 250, the selected area A user, indicated by a rectangle grid. Figure 5 presents the current frame 201 shown on the display 250, and it is shown that the same selected area V user on the display screen, as in figure 4, is used for comparison. In an embodiment in which the frame comparison before saving has been reduced in size, the current frame before the comparison operation is subjected to reduction. By reducing the size of the TV frame reduces memory required and the average frequency of occurrence of errors, which reduces the jitter of the image.

Differential relationships between the current frame and each frame of the comparison of the identifiable source (camera) is calculated in step 130. Calculates the sum of absolute values of differences between pixels of the current frame and pixels of a frame of comparison in the selected area of the user, and then divided by the sum of intensity values for pixels of the current is Adra in the selected area of the user. This ratio represents the difference between the current frame and the frame of comparison for this camera. In the first iteration, because it can be used to identify only one source, the computation of differential relations only for the exemplary data from the camera 1. The lowest value differential relationship is compared with the threshold value set by the user. If at step 140, the lowest differential value of the relationship is lower than the user threshold, the current frame is determined as coming from a source that is associated with the lowest value in a differential relationship. In one embodiment of the current frame is correlated with the source, is then copied into the frame comparison of the corresponding camera and the frame in step 150 is logged in the memory. The process of replacing the frame in comparison to the current frame facilitates the identification of the source, since this minimises the influence of gradual changes, such as changes in lighting conditions when the change of day and night and rotate the camera view.

If the minimum value of the differential relationship is established by the user threshold value or exceeds it, the user is prompted to identify the frame. The current frame and last frame of comparison for each camera is displayed on display is her user at step 170, to enable the user to visually compare frames. The current frame may be a frame from a new source or a frame from a source that has already been identified, or to submit an erroneous data that should be skipped.

If the user in step 210 identifies the frame as a frame from a new source, i.e. camera, not previously identified, the frame is copied as a frame of comparison of a new source, and the frame is written into the memory as the first frame of a new camera. If the user in step 180 identifies the frame as corresponding to an existing source, the frame is copied into the frame comparison of the corresponding source in step 150, and then outputted as the appropriate source. If the user identifies the frame as erroneous data to be skipped (step 190), the frame is discarded (step 200), and processing continues. The process is repeated from step 160 until all frames have been processed, and the operation of division is not over at step 250. It should be noted that after each camera is identified and frame comparison associated with camera-source, comparisons will be made between the current frame and the frame of the comparison of each source.

In surveillance systems, the creation of a United television image is made so that the frame or several frames from each of the second camera are stored in memory sequentially (camera 1, camera 2 and so on). As a result, after processing the first few frames usually happens identification of all cameras. To the user in this case, turning only when the frame is not consistent with any of the existing sources. Misalignment can be caused by exposure to noise, or distortion of the frame of the television image, and the data can be ignored, or can be identified the source that stores distorted or noisy frames. Misalignment can occur if the frame there are significant changes, for example, when the selected area is moving a large object.

In a specific embodiment of the present invention, the user can change the threshold. For example, increasing the threshold can be advantageous if you have excessively frequent requests to the operator when the image from the common source is not identified properly. In this embodiment, the current threshold with the current differential relations to be displayed to the user and the user can either choose one of several pre-defined threshold, or the user can enter the threshold manually.

In another embodiment, there are two different thresholds, which are compared in a differential relationship. The first threshold as described above, is used to determine the soo is vestia current image of any of the cameras. The second threshold defines the level above which frames should be dropped. If exceeded the second threshold, the data associated with the corresponding image are automatically discarded. The second threshold has a value greater than the magnitude of the first threshold may be determined based on the study of previous work experience or by experiment.

In one embodiment of the invention, a television signal from the camera 300, 301, 302, which shoots space, sequentially scan it is divided, as shown in Fig.6. In this embodiment, when setting the threshold to bind the image to the camera are taken into account temporal changes in the image in its review of the camera. For example, figure 6 shows the three cameras (camera 1, camera 2 and camera 3)generating a sequence of combined television frames 320, where this sequence is gradually added images from each camera. After the frame from all three cameras was saved as part of the sequence, the process is repeated. Figure 6 image from camera 1 appear at the moments t0 and T3. As shown in Fig.6, if the image is from time t0 will be shifted three pixels by moving the camera during the time between t0 and T3, and images are imposed, most of the pixels imaging the deposits at the time t0 overlaps with the image at the time T3. By replacing the frame in comparison to the current frame and use the updated frame of comparison in the comparison procedure minimizes the amount of displacement between the frame comparison and the current frame, so there is no need to use more sophisticated methods for tracking and comparing images. If the camera movement speed is low compared with the interval between saving images, the choice of the threshold can be adjusted to the speed of movement.

In other embodiments perform to consider moving a field of view of the camera, before performing the comparison with the current frame is the pixel offset of the frame comparison. In addition, by comparing the current image can be shifted several times in several directions when calculating the differential relations and comparing the relationship with a threshold. Further, it is possible to predict the direction of movement based on the analysis of trends, when the movement direction is determined based on the previous movements of the same source. Thus, the tracking direction, it is possible to reduce the number of shifts of the data storage cells and comparison operations.

In another specific embodiment, the threshold is set to a single cell in such a way that identifies the magic cube MOV is I or changes in the frame due to changes in the field of view of the camera, and only these images are stored in the memory cells of the television image. In this embodiment uses the same method of comparison, but the current frame and the frames of the comparison are stored in RAM, and frames that exceed the threshold are stored in the permanent memory associated with the system. In another embodiment, when the threshold is exceeded, the image is presented to the system user and the display screen is prompted to determine whether to forward the frame to be stored in permanent memory.

In some circumstances, the televised images before comparison with the current frame can be separated in time and/or date. In some systems, multiplexing, where the cameras used digital cameras, for example using a CCD (devices with charging connections), the digital data stream containing the frame in digital form, before the digital television signal frame may contain a time stamp and date. In this case, the storage and separation of personnel can be carried out in accordance with the date and time. In other systems, multiplexing, where the camera produces an analog signal with the superimposed signal date and time, or in which the digital camera enter the date and time within the digital frame, the televised images can be separated by time and date. In this embodiment, made what I'm after as the digitized video signal on the display of the displayed date and time is the definition of the selected control. If television frames must be separated by time, the selected user can be an hour 400A shown in Fig.7, and at each change in this area of the TV frame data of a television frame is stored in a new memory location associated with that hour. Similarly, given a user-selected region is the day of the month 400V, as shown hatched on the television display 410 digital image data (Fig). In this case, the system can continuously monitor changes in the date in a user-selected area exceeding the threshold, to identify frames belonging to different days of the month.

Figure 9 presents another embodiment of the invention, where the invention has the shape of the device 500. The device 500 may be an electronic device such as a module on the basis of user-programmable gate arrays (FPGA), or a specialized chip integrated circuit (ASIC). The device 500 includes a module 501 technique, which receives the digital data 550 images. The digital image data includes frames of video data from various sources. Module 501 receiving receives the first set of digital video data, predstavljajushej the first image, and also takes the current set of digital video data representing the current image. Module 501 receiving passes the first set of digital data in the module memory 502. Module memory 502 sends for storage in the storage device 560 first set of digital video data in a memory cell associated with the first source 561. These data also identified as exemplary video data of the first source. Module 501 receiving passes the first set of digital video data together with the current set of digital video data in the processing module 503. The processing module 503 calculates a differential ratio using the first set of digital video data and the current set of digital video data, as described above. If you have identified more than one source, accepted digital video data for each of the reference frame of the source, and for each, the computation of differential relations. Differential ratio(I) is sent to the module 504 requests. Module 504 query compares the differential relation(I) with a threshold. If the differential with respect to the reference frame of the source is below the threshold, then the current frame is associated with this source.

Then, the digital video data stored in the memory associated with the identified source 562. For example, if the differential with respect to exemplary ka is RA 4-th source is below a threshold, current digital video data will be associated with this source. In addition, the current frame will be an exemplary frame of the 4-th source.

If the values of all differential relations exceed the threshold, the module 504 requests sends a request to the display 510 prompting the device user to classify the current digital video information as related to the identified source or originating from a new source. In this way, the user receives a display 510, a message that the current frame of video data cannot be definitely associated with a particular source, as the differential ratio exceeds the threshold. The user can, using a device 511 input, such as a keyboard or mouse connected to the device to send a response, which will determine whether the current frame of the first source or other previously identified source, whether to associate it with the new source, or it must be discarded. Module 504 receives requests this information and instructs the module 502 memory to save the current digital video data for the current television frame in the memory cell with an identified source. The user system displays the current image, and each exemplary frame for the source. This allows the user to perform a visual comparison is s, contributing to the identification process.

It should be understood that if the user identifies the current digital video data as coming from a new source, the current digital video data will be associated with a memory cell associated with the new source. The device may also be included module 505 normalization. The normalization module normalizes the digital video data of the current television frame when compared to standard TV frames. As was shown above, the digital video data normalization allows to take into account changes in scene illumination during the day and night. Normalization is performed before the calculation of the differential relations.

It should be noted that the block diagrams are used here to illustrate various features of the invention and should not be construed as limiting the present invention to any particular logic or option run. Described logic may be divided into separate logical groups (for example, programs, modules, functions, or subroutines) without changing the overall results or any other deviation from the true scope of the claims of the invention. Often logic elements could be added, modified, removed, be performed in a different order or be carried out using other logical is their structures (for example, logic gates, a cyclic basis elements and other logical structures) without changing the overall results or any other deviation from the true scope of the claims of the invention.

The present invention can be performed in many different forms, including computer program logic for use with a processor (e.g., microprocessor, microcontroller, digital signal processor or General purpose computer), programmable logic for use with a programmable logic device (for example, user-programmable gate arrays (FPGA) or other programmable logic device, discrete components, integrated circuits (e.g., application-specific integrated circuits (ASIC)), or any other means including any combination thereof, but not limited to.

Computer program logic that performs all or part of the functions described above may be implemented in various forms, for example in the form of source code, in a form executable by a computer, and various intermediate forms (e.g., forms generated by an assembler, compiler or input device positions), but is not limited to. The source code may include a sequence of commands of computer programs implemented on any of a variety of language is in programming (e.g., object program, Assembly language or high level language such as Fortran, C, C++, JAVA, or HTML) for use with various operating systems or operating environments. The source code can identify and use different data structures and messages. The source code may be executable by the computer form (for example, through interpretive programs), or the source code may be converted (for example, by a compiler, assembler or compiler) code, executable by a computer.

The computer program may be recorded in any form (for example, in source code form, a form executable by a computer, or an intermediate form), either permanently or temporarily in the environment of the storage device (e.g., RAM, ROM, PROM, EEPROM or programmable RAM group overwriting), magnetic memory (e.g., a diskette or fixed disk), an optical storage device (e.g. CD-ROM), a PC card (e.g., card Association PCNCIA), or in another storage device. The computer program may be written in any form of signal that can be transmitted to the computer using a variety of communication methods, including analog communication, digital communication, optical communication, wireless communication, network communications and connectivity between networks. Computer programs can rasprostranyatsya in any form using the portable storage medium with accompanying documentation in printed or electronic form (for example, program, Packed in shrink film or magnetic tape)to be preinstalled in a computer system (e.g., permanent memory system or fixed disk), or distributed from a server or electronic systems teleconferencing system of communication (such as Internet or the worldwide web).

Logical devices (including programmable logic for use with programmable logic devices)that perform all or part of the functions described above can be designed using traditional methods manually, or may be designed, captured, simulated, or presented in the form of an electronic document, using different tools, such as computer-aided design (CAD), description language hardware (e.g., VHDL or AHDL), or a programming language programmable logic devices (for example, PALASM, ABEL or CUPL).

The present invention may be implemented in other specific forms within the scope of the claims of the invention. Describes the different ways of execution are not restrictive and should be considered in all respects only as an example.

1. A method for the identification of digital video data from multiple sources, wherein receiving the first set of digital video data that is presented is managing the first image, store them in the memory cell associated with the first source, which is identified by source, in the form of a standard digital video data of the first source, receive a current set of digital video data representing the current image, then calculate the differential ratio, representing the difference between the exemplary digital video data and the current set of digital video data, by calculating the difference between the pixels of the corresponding image, and when exceeding the differential relation to its specified threshold generate on the display a request for determining the membership of the current digital video data to the identified source or origin from a new source, and these sources, memory and a display connected through processor.

2. The method according to claim 1, wherein the request is sent to the display processor and upon receipt of the response to the query indicating the relation between the current set of digital video data with the new source, additionally retain the current set of digital video data in the second memory cell associated with the second source, in the form of a standard digital video data of the second source.

3. The method according to claim 1, characterized in that a differential relation below the predetermined threshold ve is icine retain the current digital video data in the form of a new model of digital video data for the first source in the memory cell, associated with the first source.

4. The method according to claim 1, characterized in that before the calculation of the differential relations additionally carry out the normalization model and the current digital video data to compensate for differences in brightness of the respective pixels of these video and expelling it from the specified calculate the value differences between pixels.

5. The method according to claim 1, characterized in that the form sets the standard digital video data for different sources and additionally carry out the calculation of a differential relationship for each set of exemplary digital video data from a source and a current set of digital video data and compare each differential relationship with his predetermined threshold value.

6. The method according to claim 5, characterized in that it further carry out the identification of the source for which is a differential relationship below its predetermined threshold value, and they identify the current set of digital video data as the reference set of digital video data for this source.

7. The method according to claim 1, characterized in that the sets of video data is subjected to compression with their reduction of the required memory, and the first set of video data before saving, and the current set of digital video data before determining a differential relationship.

8. The method according to P1, characterized in that it further distinguish the first image selected by the user part and receive the signal containing information about the relevant part of the current image, and used in the calculation of a differential relationship the same part of the current and reference images.

9. The method according to claim 1, characterized in that a differential relation below its predetermined threshold value, additionally they identify the current image as the reference image source, and the data set for the sample image used for comparisons with the data set for subsequent images.

10. The method according to claim 1, wherein when identifying each originating from different video sources, a set of data representing the current image, save as a perfect image for this source.

11. The method according to claim 1, characterized in that the sources are stationary surveillance cameras.

12. The method according to claim 1, characterized in that the sources are rotatable surveillance camera.

13. The method according to claim 1, characterized in that by means of the processor additionally take aim user signal containing information representing a predetermined threshold value specified differential relations.

14. The way the identifier is the digital video data from multiple sources, wherein receiving the first set of digital video data representing the image, store them in the memory cell associated with the first source, in the form of a standard digital video data of the first source, receive a current set of digital video data representing the current image, select a subset of the first set of digital video data, and choose a subset of the current set of digital video data, and the subset of the first set and the subset of the current set have the same pixel location, and then calculate the differential ratio, representing the difference between the subset of the first set and a subset of the current set, by calculating the difference between the pixels corresponding images and when exceeding the differential relations of its specified threshold generate on the display a request for facilities current digital video data to the identified source or origin from a new source, and these sources, a memory and a display connected through the processor.

15. The method according to 14, wherein the request is sent to the display processor and upon receipt of the response to the query indicating the relation between the current set of digital video data with the new source, additionally preserves the t current set of digital video data in the second memory cell, associated with the new source, in the form of a standard digital video data of the new source.

16. The method according to 14, characterized in that a differential relation below the predetermined threshold value keep current digital video data in the form of a new model of digital video data for the first source in the memory cell associated with the first source.

17. The method according to 14, characterized in that before the calculation of the differential relations additionally carry out the normalization model and the current digital video data to compensate for differences in brightness of the respective pixels of these video data, and excluding it from the calculation of the magnitude of the differences between pixels.

18. The method according to 14, characterized in that the form sets the standard digital video data for different sources and additionally carry out the calculation of a differential relationship for each set of exemplary digital video data from a source and a current set of digital video data and compare each differential relationship with his predetermined threshold value.

19. The method according to 14, characterized in that it further carry out the identification of the source for which is a differential relationship below its predetermined threshold value, and they identify the current set of digital video data as samples of the first set of digital video data source.

20. The method according to 14, characterized in that the sets of video data is subjected to compression with their reduction of the required memory, and the first set of video data before saving, and the current set of digital video data before determining a differential relationship.

21. The method according to 14, characterized in that a differential relation below its predetermined threshold value, additionally they identify the current image as the reference image source, and the data set for the sample image used for comparisons with the data set for subsequent images.

22. The method according to 14, wherein when identifying each originating from different video sources, a set of data representing the current image, save as a perfect image for this source.

23. The method according to 14, characterized in that the sources are stationary surveillance cameras.

24. The method according to 14, characterized in that the sources are rotatable surveillance camera.

25. The method according to 14, characterized in that by means of the processor additionally take aim user signal containing information representing the specified threshold differential relations.

26. A method for the identification of digital video data from multiple and the sources, characterized in that accept United digital video, including a number of frames, analyse and compare the digital pixel data of different frames combined digital video sequences on the basis of which each frame is identified by the source during one pass through the integrated digital video sequences.

27. The method according to p, characterized in that when comparing digital video data from the analyzed frames, the pixels of the first analyzed frame is compared with the pixels of the second analyzed frame, and if the second analyzed frame is identified as coming from the same source as the first analyzed frame, the second analyzes the frame identified as the reference frame of the source.

28. The method according to item 27, wherein the model frame used for comparison with the third analyzed by the frame and on the basis of comparison pixels in these frames determine whether the third analyzed frame from this source.

29. The TV surveillance system, characterized in that it contains a group of cameras for issuing a signal processor for receiving each of the video signal, and combining the signal in the combined signal, a storage device for storing the combined video signal, the display associated with the processor with the ability pickup is and the display request supplies the current signal, the input means associated with the processor with the processor request from a user to the division and separation of the combined video signals in a single pass on the combined signal.

30. Device for the identification of digital video data from multiple sources, characterized in that it comprises a receiving module, associated with the sources with the possibility of obtaining the first set of digital video data representing the first image, and obtaining a current set of digital video data representing the current image, a memory module for storing the first set of digital video data received from the receiving module, the memory cell associated with the first source, which is identified by the source computing module for computing a differential relations, representing the difference between the exemplary digital video data and the current set of digital video data, by calculating the difference between the pixels of the corresponding image, display and query engine that is associated with the computing module and display the query on the display and receiving a response set of digital video data to the identified source or origin from a new source in excess of a Delta against it given Rogovoy value.

31. The device according to item 30, wherein the memory module is connected with the possibility of the direction of the current set of digital video data for storage in the second memory cell associated with the second source, in accordance with the received response indicating that the current set of digital video data is associated with the new source.

32. The device according to item 30, wherein the memory module is connected with the possibility to set the current digital video data for storage in a memory cell associated with the first source when the value of the differential relations below its predetermined threshold value.

33. The device according to item 30, characterized in that it further comprises associated with the computing module normalization model and the current digital video data to compensate for differences in brightness of the respective pixels of these video data and exclude them from the calculation of the magnitude of the differences between pixels.

34. The device according to item 30, wherein the computing module is a means of calculating a differential relations for each of the reference set of digital video data representing the source, and the current set of video data, and query engine is also a means of comparing a differential relationship with a given user's threshold.

35. The device according to item 30, featuring the be fact, the module queries is also a means of identifying the source for which the magnitude of the differential relations below its predetermined threshold value, and the memory module is a means of associating the current set of digital video data with an identified source.