RU2242047C1 - Open system for automation of processing, analysis and recognition of images - Google Patents

Abstract

FIELD: computers.

SUBSTANCE: system has block 34 for task execution, block 36 for processing results analysis, display 42 and database 60 for processing results. Also provided are central block 1 for control and exchange, user interface 2, working database 10, memory block 12 for long storage, system database 20, block 30 for controlling task execution, visualization block 40 and block 50 for control and adjustment.

EFFECT: higher efficiency.

5 cl, 2 dwg

Description

The invention relates to computer science and computer engineering and can be used to automate the processing, analysis and recognition of images.

Image analysis and recognition is a priority area of computer science, applied mathematics and information technology (see, for example, A.L. Gorelik, I. B. Gurevich, V. A. Skripkin. “The current state of the recognition problem. Some aspects.” M. : Radio and communications, 1985. P.162; IB Gurevich. “The problem of image recognition” // “Recognition, classification, forecast. Mathematical methods and their application": Yearbook / Edited by Yu.I. Zhuravlev. M .: Nauka, 1989. Issue 1. P.280-329.). The tasks of image analysis and recognition are widespread in industry, medicine, automation of scientific research, ecology, geology, economics, forecasting natural disasters and emergencies, military and other industries.To solve these problems, algorithmic-software systems using modern computing tools have been created (see, for example, publications WO 00 / 13142 A1, G 06 T 5/50, 03/09/2000; WO 00/26857 A1, G 06 T 11/00, 05/11/2000; WO 01/67389 A1, G 06 T 1/00, 09/13/2001; WO 99/53445 A1, G 06 T 17/00, 10/21/1999; WO 99/63320 A3, G 06 F 17/60, 12/9/1999; RU 2159958 C1, G 06 K 9/36, G 06 T 1/00, 11/27/2000; RU 2160467 C1, G 06 K 9/00, 10.12.2000; EP 0393823 B1, G 06 T 7/20, 05.24.1995; EP 0637814 A2, G 06 T 15/10, 02/08/1995; EP 0666546 A1, G 06 T 11/00, G 06 T 11/60, 09/09/1995; EP 0681267 A2, G 06 T 1/20, 08/08/1995; EP 0725367 A1, G 06 T 15/50, 08/07/1996; RU 99113845 A, G 06 T 9/00, 04/27/2001; RU 99120046 A, G 06 K 9/46, 11/20/2001; RU 2172018 C2, G 06 T 1/60, H 04 N 7/08, 08/10/2001; RU 2173880 C2, G 06 K 9/46, 09/20/2001).

Information and computational and mathematical aspects of the problem are considered in sufficient detail (see, for example, Yu.I. Zhuravlev, I.B. Gurevich. “Pattern recognition and image recognition // Recognition, classification, forecast. Mathematical methods and their application": Yearbook Issue 2. M .: Nauka, 1989. P.5-72; IBGurevitch “The Descriptive Framework for an Image Recognition Problem" // Proceedings of The 6th Scandinavian Conference on Image Analysis, Oulu, June 19-22, 1989 : in 2 volumes, V.1: Pattern Recognition Society of Finland, 1989. - P.220-227; Yu.I. Zhuravlev, I.B. Gurevich. “Pattern recognition and image analysis" // Artificial Intelligence: 3 x nigah, Book 2: Models and Methods: Reference Book, Moscow: Radio and Communications, 1990. P.149-191; Yu.I. Zhuravlev, IB Gurevich, Methods and Means of Converting and Processing Information in Pattern Recognition Problems and image analysis // Parallel information processing: in 5 vols., vol. 5: Problem-oriented and specialized means of information processing. - Kiev: Naukova Dumka, 1990. - P.218-318).

A known method and system for identifying digital two-dimensional objects, in particular, recognition within the framework of the well-known alphabet (RU 2163394 C2, G 06 K 9/00, 02.20.2001). The method involves obtaining segmented images of objects using a computer system. A training sample is generated from the segmented image of the object using rotation and zooming operations, descriptions are generated in the form of functional dependences of the statistical estimates of the signs on the size of the segmented images of the objects, sizes are measured and normalized shape signs of the selected object are calculated. Then, the features of the form stored in the memory of the computer system are selected, and the selected object is identified based on the comparison data.

A number of developments concern the methods of adaptive recognition of information images and systems for their implementation. Thus, the invention (RU 2132061 C1, G 01 N 33/48, G 06 K 9/62, 06/20/1999) describes adaptive automatic segmentation and recognition of cells in images of cytological preparations. In the process of border segmentation, the digitized frame is divided into facets and then the boundaries of the cell structures are formed.

Also known is a device for processing color images with adaptive hybrid image filtering (RU 2159958 C1, G 06 K 9/36, G 06 T 1/00, 11/27/2000). The device comprises a control unit, a synchronization unit, neighborhood and aperture generation units, units for calculating norms, local dispersions of a useful signal, a unit for calculating filter expression values, and other functional units. The system is designed to process spectrozonal images represented by vector signals.

Another invention describes a system for adaptive recognition of information images (RU 2160467 C1, G 06 K 9/00, 12.10.2000), which although it improves image quality, but has rigidly fixed valid basic image processing functions.

A known method and system for constructing means for image processing (WO 01/67389 A1, G 06 T 1/00, 09/13/2001), which contains means for inputting images, storing, image processing and visualization. This system is a multifunctional image analysis tool designed for application development, research and training. The system includes

- the core of the system, which consists of modules that invoke operators, control input / output operations and translate data objects into the internal language of the system;

- data management tools that are responsible for managing the main memory (optimized for image processing), as well as for creating, managing and destroying graphic objects and tuples;

- means of algorithmic filling - a library including more than 750 operators for image analysis and data visualization;

- HDevelop tool for accelerated development of image analysis programs (Computer Aided Vision Engineering CAVE - tool);

- debugging tools;

- a knowledge base of operators, which contains reference information for each operator of the system (names and number of operators, parameter types, proposed parameter values).

This is an open system in which two main programming methods are supported: a) interactive development of programs using HDevelop; b) using external C and C ++ language compilers (there is an interface for these languages).

This system is the closest analogue of the patented system, however, it has disadvantages, which boil down to the following:

a) Plug-in technology is not supported for loading visualization interfaces and editing new data types entered by the user;

b) requires the presence on a computer operating systems of a level not lower than Windows NT 4.0;

c) the possibility of simultaneous execution of several tasks inside the shell is not provided (there is no multithreaded executing subsystem);

d) there are no means of editing information about the algorithms of the database set, which does not allow to add additional information about the algorithms.

In addition, the disadvantage of the closest analogue, like the other systems mentioned above, is that they are “closed” or “quasi-open”, since image processing and analysis algorithms are specified during system design and changing a given set of algorithms is either not provided or requires high user qualifications. This fact does not allow or significantly complicates the addition or replacement of new algorithms in the process of image processing. In addition, the purpose of known systems is to solve applied problems, but not to develop and test new algorithms for solving applied problems of analysis and evaluation of information presented in the form of images.

The objective of the invention is to build an open system for automating the development, research and testing of algorithms for processing, analyzing and recognizing information presented in the form of images, as well as automating the solution of applied decision-making problems in which the source information is presented in the form of images of an arbitrary nature and character.

EFFECT: expansion of the nomenclature of task classes and types of processed objects, including images of arbitrary type, expansion of a set of processing algorithms and analysis of images and drives for data visualization.

The technical result is ensured by the fact that an open system for automating the processing, analysis and recognition of images includes a task execution unit associated with its outputs with a database of processing results and with an analysis unit for processing results, algorithmic filling tools, visualization tools with a display. A central control and exchange unit, a user interface, a working database, a long-term storage unit, a system database, a task execution control unit, a visualization unit and a control and adjustment unit connected by bidirectional control and information signal transmission buses with a control input are introduced into the system: the output of the database of processing results and with the first input-output of the Central control and exchange unit, the second, third, fourth, fifth and sixth inputs and outputs of which are connected bi-directional buses, respectively, with the first output-input of the job execution control unit, the output-input of the system database, the output-input of the visualization unit, the output-input of the user interface and the first output-input of the working database. The second input-output of the working database is connected to the control output of the storage unit for long-term storage, the information output of which is connected to the information input-output of the working database associated with its output with the first input of the visualization unit, the output of which is connected to the display input, and the second and third inputs - to the output of the system database and, respectively, to the third output of the task execution unit, the first output of which is connected to the input of the processing results analysis unit, and the second - to the database input yes data of processing associated with its output with the input of the storage unit for long-term storage. The output of the processing results analysis unit is connected to the input of the central control and exchange unit, the control output of which is connected to the control input of the job control unit, connected by its control output to the control input, and the second output-input to the input-output of the job execution unit.

The system can be characterized by the fact that the control unit for the execution of tasks contains a block for generating a task for image processing and a block for a bank of algorithmic modules, the control input of which is connected to the first control output of the block for generating a task for image processing, and the output-input is the second output-input of the control execution of tasks, the first output-input and control input of which are the output-input and control input of the image processing task forming unit, respectively A swarm control output of which is the control output of the job execution control unit.

The system may also be characterized in that the system database contains an interface unit of the system database, a system database manager, an algorithm database, a program module database, a parameter type database and a directory database, the first output-input and output of the interface unit are respectively the output-input and output of the system database, and the second output-input of the interface unit is connected to the first input-output of the system database manager, the second, third, fourth and fifth inputs and outputs of which connected by separate bi-directional buses with outputs-inputs, respectively, of the database of algorithms, the database of program modules, the database of parameter types and the database of catalogs.

The system can be characterized by the fact that the working database contains an interface unit of the working database and sequentially connected by bidirectional buses, a storage unit, a storage unit for data rollback and a storage unit for deep rollback, the control inputs of which are connected to the control output of the interface unit, the first output the input and output of which are, respectively, the first output-input and output of the working database, the second input-output of which is the second output-input of the interface unit, the third output-input of which is connected with the first information input-output of the storage unit, the second information input-output of which is the information input-output of the working database.

The system can be characterized, in addition, by the fact that the database of processing results contains an interface block of a variable structure and a storage block of data of a processing result, the input of which is an input to the database of processing results, the output of the interface block of a variable structure is the output of a database of processing results, and control input-output - control input-output of the database of processing results.

The basis of the construction of the patented system are the following provisions. Image processing and recognition systems, as mentioned above, are built on the principle of closedness of a set of admissible transformations, i.e. processing algorithms are predefined and cannot be changed by the user during the operation of the system. The patented system is based on a different principle of image analysis and processing - the openness and adaptability of a set of permissible transformations. The foundations of the methodology for constructing such systems are laid in the work of Yu.Zhuravlev, I.B. Gurevitch, S.V. Ilyinski et al. "Development and Investigation of the Mathematical and Computational Basis for a System of Information Technologies of Pattern Recognition and Image Understanding" // Pattern Recognition and Image Analysis: Advances in Mathematical Theory and Applications. 1993, v.3, N 3., pp.266-282, and developed in the work of I.B. Gurevich, Yu.I. Zhuravlev, D.M. Murashov et al. “A system for automating scientific research in the field of analysis and understanding images based on the accumulation and use of knowledge. Part 1 ”// Autometry. - 1999. - No. 6. - S.23-50. In accordance with this methodology, the transformation algorithm (processing, analysis, recognition) is constructed by means of a software package according to the block-structural principle. If the desired block is not in the structure of the algorithm, it can be added or constructed using the tools of the patented system itself (macros or scripts). This allows the user to expand the set of image processing and analysis algorithms in the algorithmic bank and replenish information for their search in databases. The same method is used to expand the sets of types of processed data. It is possible, for example, to add a pyramidal representation of an image or an image with a “legend” - a semantic description that can be used by the processing algorithm. Expanding the set of drives for data visualization greatly facilitates the processing, multidimensional analysis, recognition and classification of images of an arbitrary nature.

The theoretical basis of the algorithmic filling of the system and scenarios available to the user is a descriptive approach to the analysis and recognition of images (I. Gurevich “The problem of image recognition” // “Recognition, classification, forecast. Mathematical methods and their application": Yearbook / Ed. Yu.I. Zhuravleva, Moscow: Nauka, 1989. Issue 1. P.280-329; IBGurevitch “The Descriptive Framework for an Image Recognition Problem" // Proceedings of The 6th Scandinavian Conference on Image Analysis, Oulu, June 19-22, 1989: in 2 volumes, V.1: Pattern Recognition Society of Finland, 1989. - P.220-227; IBGurevitch “Descriptive Technique for Image Description, Representation and Recognition "// Pattern Recognition and Image Analysis: Advances in Mathematical Theory and Applications in the USSR. 1991. V.1, N 1. P.50-53). Accordingly, processing, transformation, analysis and recognition of images in the system are implemented through the following the main methods provided by the descriptive approach: a) standardization of descriptions of image processing and analysis tasks; b) standardization of descriptions of image processing and analysis procedures; c) the use of a single formal system for describing and implementing operations on image processing and analysis procedures and image models (image algebras and descriptive image algebras); d) the use of combinations of basic image conversion procedures; e) reduction of images to a form convenient for recognition (synthesis and formalization of image descriptions, presentation of images by “signs” and models, reproduction in the description of the structure, semantic and spatial relations of the image); f) the implementation of image recognition based on identification methods, assessing the proximity of images in the spaces of descriptions and matching representations of images; g) using the properties of invariance and equivalence of images.

The invention is illustrated in the drawings, where Fig. 1 is a block diagram of a patented system; figure 2 is a view of the shell window of the system.

An open system for automating the processing, analysis and recognition of images (Fig. 1) includes a central control and exchange unit 1 connected to a user interface 2, a working database 10, a system database 20, a block 30 for executing tasks, a block 40 for visualization, a block 50 controls and settings.

Working database 10 includes an interface block 11 of the working database, which provides data visualization management, data management and storage, as well as the formation of backup data arrays for rollback. The working database 10 has in its composition storage units 14, 16 and 18.

The interface unit 11 of the working database is connected with the storage unit 12 for long-term storage, which is connected with the storage unit 14. The interface unit 11 of the working database is also connected with the unit 14. The functions of the storage unit 14 are to store data, for example, in a dedicated area of mainframe memory and in providing access to data. The storage unit 14 also provides a count of the number issued for use links. The storage unit 14 is associated with the storage units 16 of data for rollback and 18 data for deep rollback. The data storage unit 16 for rollback provides a backup data unit for rollback, and the data unit 18 provides a backup data unit for deep rollback.

The Central unit 1 control and exchange is connected with the system database 20 data. The system database 20 comprises an interface unit 21 of a system database associated with a system database manager 22. The system database manager 22 is associated with a database of 23 algorithms, a database of 24 software modules, a database of 25 parameter types, a database of 26 catalogs.

The central control and exchange unit 1 is also connected to the task execution control unit 30. Block 30 comprises a block 32 for generating an image processing task associated with a task execution unit 34, which is associated with a processing result analysis unit 36. The output of the analysis unit 36 is connected to the central control and exchange unit 1. Blocks 32 and 34, in addition, are associated with block 38 of the bank of algorithmic modules.

The central control and exchange unit 1 is connected to the visualization unit 40, to the other inputs of which the interface unit 11 of the working database, the interface unit 21 of the system database, the unit 34 for executing the task are connected, and the display 42 to the output.

The central control and exchange unit 1 is connected to the control and adjustment unit 50, which is connected to the processing result data base 60, which includes an interface unit 62 of variable structure and an associated storage unit 64 of the processing result data. The variable structure interface unit 62 is also connected to the long-term storage unit 12.

The functions of the central control and exchange unit 1 include managing a working database 10 and a database 60 of processing results, i.e. issuing a command to create a new data and service information array and closing an existing and registered data and service information array, transmitting commands from user interface 2 to any interface unit, maintaining a list of created blocks 11, 21 and 62 - drives for data visualization.

Further, the central control and exchange unit 1 transfers commands from one drive to another, transfers commands for changing the image formation order from the user interface 2 to the visualization unit 40, and generates a list of drives satisfying the request condition.

The functions of the long-term storage unit 12 are to provide long-term storage of the data block-file, as well as to provide data retrieval upon request from the interface unit 11 of the working database and from the control and adjustment unit 50.

To ensure the functioning of the databases 23-26 included in the system database 20, the interface unit 21 and the dispatcher 22 of the system database are intended. The interface unit 21 and the manager 22 of the system database database provide receiving commands for editing databases, as well as visualization of information from the databases. The system database manager 22 performs input and output of database information, searches for information on demand, supports semantic relationships between databases, and also provides transactions.

Databases included in the system database 20 provide the following functions:

- a database of 23 algorithms data provides storage of information about algorithms and functions, adding, deleting, modifying records and organizing them, as well as searching for records;

- a database of 24 data of program modules provides storage of information about registered program modules, adding, deleting, changing records and organizing them, as well as searching for records;

- a database 25 of data of parameter types provides storage of information about the types of parameters that can be used, and accordingly add, delete, modify records and their ordering, as well as search records;

- the database 26 of the catalogs supports the tree-like catalog of algorithms, as well as adding, deleting, changing records and their ordering and, in addition, searching for records.

Unit 30 controls the execution of tasks provides the formation of tasks for image processing, the execution of such tasks in multi-threaded mode and analysis of its results. It can work in 2 modes - debugging and regular. Accordingly, the image processing task generating unit 32 provides the following functions: analyzes the information about the algorithm, compiles lists of parameters according to information from the control unit, compiles a list for selecting the values of each parameter, selects parameter values by command from the user interface 2. At the same time, when selecting parameter values at a command from the central control and exchange unit 1, block 11 blocks the storage unit 14 from sharing and copies the contents of block 14 to the storage unit 16, and block 16 to the block 18 to enable rollback of the operation. If the contents of block 14 were not modified and the data was not processed, then blocks 16 and 18 may not contain data. If the data from the working database 10 data has not yet been requested for processing, then blocks 16 and 18 do not contain information (empty). In addition, block 32 downloads the necessary algorithmic module and searches for the necessary function, issues commands to block data, and checks the correctness of the choice of parameters. Block 32 transfers to the block 34 of the job job the address of the function, data, links to data blocks, and also generates actions for a possible refusal to run the job.

Unit 34 execution of the task, respectively, provides the execution of the task, changing its priority, as well as a reflection of the degree of completion. Block 36 analysis of the results of processing analyzes the results of the execution of the task of block 34, issuing commands to unlock the data and create a new data container by block 50.

The block 30 for controlling the execution of tasks includes a block 38 of a bank of algorithmic modules, which provides storage of algorithmic modules and search for modules upon request from block 32 for generating an image processing task.

The functions of the control and adjustment unit 50 consist in forming a new data and service information array in the data base 60 of the processing results through the interface unit 62 of the variable structure, as well as the corresponding preparation of the system database 20 and the interface unit 21 of the system database. The functions of the variable structure interface unit 62 provide for the visualization, management and storage of data, as well as the formation of redundant data blocks for rollback.

The interface unit 62 of a variable structure (like block 11) is intended for storing data in a allocated area of mainframe memory, providing access to data, freeing up random access memory occupied by data, and also for counting the number of issued links for use.

Block 40 visualization supports the formation of images on the screen of the display 42, the imposition in the desired order of images from various interfaces of the system, as well as changing the order of image formation on the screen of the display 42 by command from the central unit 1 of the control and exchange.

The patented system can be implemented in the form of a multi-machine complex, depending on the requirements for system performance, since it allows the parallel execution of operations and actions. With low performance requirements, the system or its individual parts can be implemented on several or on one personal computer. Patented structural principles allow the use of modern principles of program design for implementation on one personal computer. The performance of the patented system is confirmed by the pilot operation of its software model, designed for use in the Windows environment, which allows the use of OLE technology for linking and embedding objects, while blocks such as 11, 32, and 62 are respectively represented as OLE drives.

The system operates as follows.

An image, or a numerical matrix, or a sequence of images, to be processed and analyzed by the user’s choice, is read out from the storage unit 12 for long-term storage to the storage unit 14, in which, using protocols, by the command of the central control and exchange unit 1 of the working database data coordination the contents of the working database 10 data is formed for this image.

Next, the central control and exchange unit 1 issues a command to the system database 20 to select the necessary algorithm for processing the image stored in block 14. The algorithm request is generated through the interface unit 21 of the system database and the dispatcher 22 from the databases 23, 24, 25 and 26. The algorithm is selected from the catalog, located by keyword or more complex query.

Information about the algorithm found through the central control and exchange unit 1 is supplied to the task execution control unit 30 and transmitted to the image processing task generating unit 32. Block 32 provides the user with a choice of algorithm parameters from prepared lists of possible options. These lists are formed by the central control and exchange unit 1 according to information about the type of parameter and additional restrictions on the properties of this parameter. So, for example, if the number of neighbors (“connectivity”) of each pixel is indicated, then only three options are possible: 4, 6 and 8. Block 32 captures the user's choice, blocks the data used from unauthorized access (from another task), finds the program module in block 38 of the bank of algorithmic modules and transmits the task for execution to block 34. The block 34 of the task execution controls the execution of the task and allows to increase or decrease its priority (several tasks can be executed at the same time), shows the degree of completion of the task I. After completion of the task, block 34 transfers the results to block 36 analyzing the results of processing.

Block 36 analyzes the result of the end of the algorithm and sends a command to release the locked data. If the result of the work is the new contents of the storage unit 64 of the data of the processing result, then the analysis unit 36 transmits a command to the control and adjustment unit 50.

To this end, the control and adjustment unit 50 from the data storage unit 64 of the processing result requests information about the data type, according to which the central control and exchange unit 1 selects the settings of the variable structure interface unit 62, which subsequently creates a new array of data and service information in the database 60 data processing results. If necessary, this array in the database 60 of the data of the processing results can be used in the next task or stored in the storage unit 12 for long-term storage. The result of the work is both the processed image and the sequence of operations (algorithms, tasks) used to synthesize this image. This sequence can be stored in the system database 20, by means of the interface unit 21 and the system database manager 22 in the respective databases as a new processing algorithm added to the set of valid transformations.

Block 40 visualization in the process of functioning of the system allows you to generate the resulting view of the data set for display 42. The resulting view is formed as a set of windows, each of which contains one of the results of the data visualization or the result of the work of the service drive (for example, information). A typical view of the shell window of a system implemented in an operating system of the Windows family is shown in FIG. 2. It can be seen that the upper part of the screen is a menu and a toolbar for managing the active window, in the central part of the screen there are panels of the system database 20, task execution unit 34, as well as a number of images visualized graphically and in the form of a table of numbers (bases 10 and 60 data).

The system with the help of object libraries (block 38 of the bank of algorithmic modules) and additional information in the system database 20 (databases 23, 24, 25 and 26) supports reading, distribution in memory, saving, copying with the following types of objects: vectors (integer numbers, real floating-point numbers), matrices (integers, real floating-point numbers, structural), images (1-, 4-, 8-, 16-, 24-, 32-bit / pixel) in BMP formats, TIFF, PCX, JPEG (reading and saving for TIFF, PCX, JPEG), fragments of vectors and matrices. The system includes, in particular, the following basic tools for image processing: affine transforms, Fourier transforms, Gabor transforms, wavelet transforms; matrix filtering, separation of brightness transitions and contours, segmentation methods, morphological operations, texture analysis methods, factor analysis algorithms, recognition and decision-making algorithms.

Using plug-in technology, new algorithms for processing, converting, analyzing and recognizing images, visualization tools, and advanced database search tools can be added to the system during operation. This is done without changing the basis of the system (control connections and architecture) while preserving the already accumulated knowledge in the databases (their number can also be increased). For example, a successful sequence of image analysis algorithms can be added as a new algorithm with a description indicating recommended areas of competence.

Thus, in the process of using the system, the number of algorithms grows, the information about them in the databases also expands, new descriptions are needed to search and use the algorithms, which extends the capabilities of the system.

The patented object is implemented in the open system “Black Square. Version 1.1 ”, designed to automate scientific research in the field of processing, analysis and recognition of images based on PC.

Claims (5)

1. An open system for automating processing, analysis and recognition of images, including a task execution unit, connected with its outputs to a database of processing results and to a unit for analyzing processing results, visualization tools with a display, characterized in that a central control and exchange unit is introduced into it , user interface, working database, long-term storage unit, system database, task execution control unit, visualization unit and control and tuning unit directional transmission buses of control and information signals with a control input-output of the database of processing results and with the first input-output of the central control and exchange unit, the second, third, fourth, fifth and sixth inputs and outputs of which are connected by separate bi-directional buses, respectively, with the first output the input of the control unit for the execution of tasks, the output-input of the system database, the output-input of the visualization block, the output-input of the user interface and the first output-input of the working database o, the second input-output of which is connected to the control output of the storage unit for long-term storage, the information output of which is connected to the information input-output of the working database, connected by its output to the first input of the visualization unit, the output of which is connected to the display input, and the second and third inputs - to the output of the system database and, accordingly, to the third output of the task execution unit, the first output of which is connected to the input of the processing results analysis unit, and the second - to the database input the processing results associated with its output with the input of the long-term storage unit, the output of the processing results analysis unit is connected to the input of the central control and exchange unit, the control output of which is connected to the control input of the job control unit connected to its control output by the control input, and by the second output -input - to the input-output block of the job. 2. The system according to claim 1, characterized in that the task execution control unit comprises an image processing task formation unit and an algorithm module bank unit, the control input of which is connected to the first control output of the image processing task formation unit, and the output-input is the second the output-input of the job execution control unit, the first output-input and control input of which are the output-input and control input of the image processing task forming unit, the second whose control output is the control output of the job execution control unit. 3. The system according to claim 1, characterized in that the system database comprises an interface unit of the system database, a system database manager, an algorithm database, a program module database, a parameter types database and a directory database, a first output-input, and the output of the interface unit is respectively the output-input and output of the system database, and the second output-input of the interface unit is connected to the first input-output of the system database manager, the second, third, fourth and fifth inputs and outputs of which are connected separate bi-directional buses with outputs-inputs respectively of a database of algorithms, a database of software modules, a database of parameter types, and a database of catalogs. 4. The system according to claim 1, characterized in that the working database comprises an interface unit of the working database and a memory unit, a data unit for rollback and a data unit for deep rollback connected in series with bi-directional buses, the control inputs of which are connected to the control output of the interface block, the first output-input and output of which are respectively the first output-input and output of the working database, the second input-output of which is the second output-input of the interface unit, t the second output-input of which is connected with the first information input-output of the storage unit, the second information input-output of which is the information input-output of the working database. 5. The system according to any one of paragraphs. 1-4, characterized in that the database of processing results contains a variable structure interface unit interconnected by a bi-directional bus and a processing result data block, the input of which is the input of the processing result database, the output of the variable structure interface unit is the output of the processing result database, and the control input-output - controlling input-output of the database of processing results.

Images