Encoding device, decoding device, encoding method, decoding method, encoding program, and decoding program

Abstract

The present invention makes it possible to generate useful decoded data even when it is not possible to set different compression rates for respective regions. This encoding device includes: a determination unit that, on the basis of a result of recognition processing by AI, determines a target region that is necessary to recognize a recognition target in image data and a threshold compression rate at which the recognition target can be recognized; a first encoding unit that encodes the entire region of the image data at the threshold compression rate and transmits the encoded image data; a generation unit that generates first invalidated image data obtained by invalidating regions other than the target region of the image data and generates second invalidated image data obtained by invalidating regions other than the target region of decoded data obtained by encoding the image data at the threshold compression rate and then decoding the encoded image data; and a second encoding unit that encodes differential data between the first invalidated image data and the second invalidated image data at a predetermined compression rate and transmits the encoded differential data.

Description

[Technical Field]

[0001] The present invention relates to an encoding device, a decoding device, an encoding method, a decoding method, an encoding program, and a decoding program. [Background technology]

[0002] Generally, when recording or transmitting image data, the data size is reduced by encoding, thereby reducing the recording or transmission costs.

[0003] On the other hand, when recording or transmitting image data for use in recognition processing by AI (Artificial Intelligence), one possible method is to increase the compression rate of each area to the limit at which the AI ​​can recognize the target (i.e., encode it at the limit compression rate). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-034342 [Patent Document 2] Japanese Patent Publication No. 2021-034983 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the case of an encoding device that cannot set a different compression rate for each region (that is, the same compression rate is set for all regions), the encoding method described above cannot be applied.

[0006] In response to this, a method can be considered in which the data size of the encoded data is reduced by, for example, blacking out areas other than the target area necessary for recognizing the recognition target, and then encoding the entire area at the limit compression rate.

[0007] However, in this method, areas other than the target area in the decoded data cannot be used as image data, which reduces the usability of the decoded data.

[0008] In one aspect, an object of the present invention is to enable generation of highly convenient decoded data even when it is not possible to set different compression rates for each region. [Means for solving the problem]

[0009] According to one aspect, an encoding device comprises: A determination unit that determines a target area required to recognize a recognition target in image data and a limit compression rate at which the recognition target can be recognized based on the result of the recognition processing by AI; a first encoding unit that encodes the entire area of ​​the image data at the limit compression rate and transmits the encoded data; a generating unit that generates first invalidated image data obtained by invalidating an area other than the target area of ​​the image data, and second invalidated image data obtained by invalidating an area other than the target area of ​​decoded data obtained by encoding the image data at the limit compression rate and then decoding the image data; The image processing device further includes a second encoding unit that encodes differential data between the first invalidated image data and the second invalidated image data at a predetermined compression rate and transmits the encoded differential data. [Effects of the Invention]

[0010] Even if it is not possible to set different compression rates for different regions, highly convenient decoded data can be generated. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating an example of a system configuration of an image processing system. [Figure 2] FIG. 2 is a diagram illustrating an example of the hardware configuration of the image processing device and the server device. [Figure 3] FIG. 3 is a first diagram illustrating an example of the functional configuration of the hierarchical encoding unit of the image processing device. [Figure 4]FIG. 4 is a second diagram illustrating an example of the functional configuration of the hierarchical encoding unit of the image processing device. [Figure 5] FIG. 5 is a first diagram showing a specific example of processing by the invalid area image processing unit. [Figure 6] FIG. 6 is a first flowchart showing the flow of the hierarchical encoding process. [Figure 7] FIG. 7 is a first diagram illustrating an example of the functional configuration of the hierarchical decoding unit of the server device. [Figure 8] FIG. 8 is a diagram illustrating a specific example of processing by the addition unit. [Figure 9] FIG. 9 is a first flowchart showing the flow of the hierarchical decoding process. [Figure 10] FIG. 10 is a third diagram illustrating an example of the functional configuration of the hierarchical encoding unit of the image processing device. [Figure 11] FIG. 11 is a second flowchart showing the flow of the hierarchical encoding process. [Figure 12] FIG. 12 is a fourth diagram illustrating an example of the functional configuration of the hierarchical encoding unit of the image processing device. [Figure 13] FIG. 13 is a fifth diagram illustrating an example of the functional configuration of the hierarchical encoding unit of the image processing device. [Figure 14] FIG. 14 is a sixth diagram illustrating an example of the functional configuration of the hierarchical encoding unit of the image processing device. [Figure 15] FIG. 15 is a second diagram showing a specific example of the processing by the invalid area image processing unit. [Figure 16] FIG. 16 is a third flowchart showing the flow of the hierarchical encoding process. [Figure 17] FIG. 17 is a second diagram illustrating an example of the functional configuration of the hierarchical decoding unit of the server device. [Figure 18] FIG. 18 is a diagram illustrating a specific example of the processing of the replacement unit. [Figure 19] FIG. 19 is a second flowchart showing the flow of the hierarchical decoding process. [Figure 20] FIG. 20 is a seventh diagram illustrating an example of the functional configuration of the hierarchical encoding unit of the image processing device. [Figure 21]FIG. 21 is an eighth diagram illustrating an example of the functional configuration of the hierarchical encoding unit of the image processing device. [Figure 22] FIG. 22 is a fourth flowchart showing the flow of the hierarchical encoding process. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, each embodiment will be described with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0013] [First embodiment] <Image processing system configuration> First, the system configuration of an image processing system including an encoding device and a decoding device will be described. Fig. 1 is a diagram showing an example of the system configuration of an image processing system. As shown in Fig. 1, the image processing system 100 includes an imaging device 110, an image processing device 120 which is an example of an encoding device, and a server device 130 which is an example of a decoding device. In the image processing system 100, the image processing device 120 and the server device 130 are connected to each other so as to be able to communicate with each other via a network (not shown).

[0014] The imaging device 110 captures images at a predetermined frame rate and transmits the moving image data to the image processing device 120.

[0015] An encoding program is installed in the image processing device 120, and the image processing device 120 functions as a hierarchical encoding unit 121 by executing the encoding program.

[0016] The hierarchical coding unit 121 encodes the image data of each frame included in the video data to generate first encoded data. When generating the first encoded data, the hierarchical coding unit 121 increases the compression rate to the limit at which the AI ​​can recognize the recognition target included in the image data, and encodes the entire area of ​​the image data at the same compression rate. In other words, the hierarchical coding unit 121 encodes the entire area of ​​the image data at the limit compression rate.

[0017] In addition, in order to compensate for the difference in image quality of the target region between the decoded data when the generated first encoded data is decoded and the image data before encoding, the hierarchical encoding unit 121 encodes the differential data between the two data to generate second encoded data. Note that the target region refers to the region in the image data that is necessary for the AI ​​to recognize the recognition target.

[0018] Furthermore, the hierarchical encoding unit 121 transmits the generated first encoded data and second encoded data to the server device 130.

[0019] A decoding program is installed in the server device 130, and the server device 130 functions as a hierarchical decoding unit 131 by executing the decoding program.

[0020] The hierarchical decoding unit 131 decodes the first coded data and second coded data transmitted from the image processing device 120 to generate first decoded data and second decoded data.

[0021] Furthermore, the hierarchical decoding unit 131 adds the second decoded data to the generated first decoded data, and stores the decoded data after the addition in the decoded data storage unit 132.

[0022] This allows the server device 130 to compensate for the degradation in image quality of the decoded data for the target region caused by encoding at the limit compression rate. Therefore, even if the image processing device 120 encodes image data at the limit compression rate suitable for AI recognition processing, the server device 130 can generate decoded data for the target region that reproduces the image quality of the image data before encoding.

[0023] Furthermore, the server device 130 can generate decoded data by decoding the first encoded data that has been encoded at the limit compression rate without undergoing processing such as blacking out. Therefore, the server device 130 can generate decoded data with improved image quality for areas other than the target area, compared to when image data that has undergone processing such as blacking out is encoded.

[0024] As a result, according to this embodiment, it becomes possible to use the decoded data for purposes other than AI recognition processing. In other words, according to this embodiment, even if it is not possible to set different compression rates for each region, it is possible to generate highly convenient decoded data.

[0025] <Hardware configuration of image processing device and server device> Next, a description will be given of the hardware configuration of the image processing device 120 and the server device 130. Fig. 2 is a diagram showing an example of the hardware configuration of the image processing device and the server device.

[0026] 2A is a diagram showing an example of the hardware configuration of an image processing device 120. The image processing device 120 has a processor 201, a memory 202, an auxiliary storage device 203, an I / F (Interface) device 204, a communication device 205, and a drive device 206. The hardware components of the image processing device 120 are connected to each other via a bus 207.

[0027] The processor 201 has various arithmetic devices such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), etc. The processor 201 reads various programs (for example, an encoding program, etc.) into the memory 202 and executes them.

[0028] The memory 202 has a main storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The processor 201 and the memory 202 form a so-called computer, and the processor 201 executes various programs read onto the memory 202, causing the computer to realize various functions.

[0029] The auxiliary storage device 203 stores various programs and various data used when the processor 201 executes the various programs.

[0030] The I / F device 204 is a connection device that connects the image processing device 120 with the imaging device 110, which is an example of an external device.

[0031] The communication device 205 is a communication device for communicating with the server device 130 via a network.

[0032] The drive device 206 is a device for loading a recording medium 210. The recording medium 210 here includes media that record information optically, electrically, or magnetically, such as a CD-ROM, a flexible disk, a magneto-optical disk, etc. The recording medium 210 may also include semiconductor memory that records information electrically, such as a ROM, a flash memory, etc.

[0033] The various programs to be installed in the auxiliary storage device 203 are installed, for example, by setting the distributed recording medium 210 in the drive device 206 and reading the various programs recorded on the recording medium 210 by the drive device 206. Alternatively, the various programs to be installed in the auxiliary storage device 203 may be installed by being downloaded from a network via the communication device 205.

[0034] 2B is a diagram showing an example of the hardware configuration of the server device 130. Note that the hardware configuration of the server device 130 is generally the same as the hardware configuration of the image processing device 120, and therefore the following description will focus on the differences from the image processing device 120.

[0035] The processor 221 reads, for example, a decoding program into the memory 222 and executes it.

[0036] The I / F device 224 accepts operations for the server device 130 via an operation device 231. The I / F device 224 also outputs the results of processing by the server device 130 and displays them via a display device 232. The communication device 225 also communicates with the image processing device 120 via a network.

[0037] <Functional configuration of the hierarchical coding unit of the image processing device> Next, the functional configuration of the hierarchical encoding unit 121 of the image processing device 120 will be described with reference to Fig. 3 and Fig. 4. Fig. 3 and Fig. 4 are first and second diagrams showing an example of the functional configuration of the hierarchical encoding unit of the image processing device.

[0038] Here, the functions of each unit when encoding one frame of image data out of the image data of each frame included in video data will be explained using FIG. 3 and FIG. 4 according to the execution order.

[0039] As shown in FIGS. 3 and 4, the hierarchical coding unit 121 includes an encoding unit 310, a decoding unit 320, a compression rate determination unit 330, a target region detection unit 340, an invalid region image processing unit 350, and an encoding unit 360.

[0040] First, the function of each unit up to the determination of the limit compression rate used when encoding one frame of image data will be described with reference to FIG.

[0041] The encoding unit 310 acquires one frame of image data from the video data, encodes the acquired image data using the compression rate notified by the compression rate determination unit 330, and notifies the decoding unit 320 of the encoded data.

[0042] The decoding unit 320 decodes the coded data notified by the encoding unit 310 to generate decoded data, and notifies the compression rate determination unit 330 of the generated decoded data.

[0043] The compression rate determination unit 330 is an example of a determination unit. The compression rate determination unit 330 performs AI recognition processing on the decoded data notified by the decoding unit 320, and determines whether the recognition target included in the image data can be recognized. If the compression rate determination unit 330 determines that the recognition target can be recognized, it increases the compression rate by a predetermined increment and notifies the encoding unit 310.

[0044] If the compression rate determination unit 330 determines that the recognition target cannot be recognized, it determines the compression rate previously notified to the encoding unit 310 as the limit compression rate. Furthermore, the compression rate determination unit 330 notifies the encoding unit 310 and the encoding unit 360 of the determined limit compression rate.

[0045] In this way, the encoding unit 310, decoding unit 320, and compression rate determination unit 330 determine the limit compression rate by repeating the cycle of encoding → decoding → recognition processing → change of compression rate.

[0046] Next, the functions of each unit up to the generation of the first coded data and the second coded data using the determined limit compression rate will be described with reference to FIG.

[0047] The target area detection unit 340 is an example of a determination unit. The target area detection unit 340 performs AI recognition processing on one frame of image data acquired from video data, and detects a target area in the image data that is necessary to recognize the recognition target. The target area detection unit 340 also calculates the position of the detected target area within the image data, and notifies the invalid area image processing unit 350 of the position information.

[0048] The invalid area image processing unit 350 is an example of a generating unit. After the encoding unit 310 encodes one frame of image data acquired from video data at a limit compression rate, the invalid area image processing unit 350 acquires the decoded data when the decoding unit 320 decodes the first encoded data.

[0049] The invalid area image processing unit 350 also invalidates areas of one frame of image data acquired from the video data other than the target area identified by the notified position information, thereby generating first invalidated image data. The invalid area image processing unit 350 also invalidates areas of the decoded data notified by the decoding unit 320 other than the target area identified by the notified position information, thereby generating second invalidated image data. The invalid area image processing unit 350 also calculates the difference between the first invalidated image data and the second invalidated image data, thereby generating difference data. The invalid area image processing unit 350 then notifies the encoding unit 360 of the generated difference data.

[0050] Note that invalidation refers to changing the pixel values ​​of each color component in an area other than the target area specified by the position information to a predetermined pixel value.

[0051] The encoding unit 310 is an example of a first encoding unit. The encoding unit 310 generates first encoded data by encoding the entire area of ​​one frame of image data acquired from the video data at the determined limit compression rate. The encoding unit 310 also transmits the generated first encoded data to the server device 130.

[0052] The encoding unit 360 is an example of a second encoding unit. The encoding unit 360 generates second encoded data by encoding the difference data notified by the invalid area image processing unit 350 at a predetermined compression rate. The encoding unit 360 also transmits the generated second encoded data to the server device 130.

[0053] In this way, the hierarchical coding unit 121 encodes the entire area of ​​the image data at the limit compression rate, and encodes the difference data indicating the degradation of the image quality of the target area at a predetermined compression rate. As a result, the hierarchical coding unit 121 can reproduce the image quality of the image data before encoding for the target area, and can improve the image quality of areas other than the target area compared to conventional methods.

[0054] As a result, it becomes possible to use the decoded data for purposes other than AI recognition processing. In other words, the hierarchical coding unit 121 makes it possible to decode highly convenient decoded data even when it is not possible to set different compression rates for each region.

[0055] <Specific example of processing by the invalid area image processing unit> Next, a description will be given of a specific example of the processing by the invalid area image processing unit 350. Fig. 5 is a first diagram showing a specific example of the processing by the invalid area image processing unit.

[0056] 5, image data 501 is image data for one frame obtained from video data. As shown in Fig. 5, the invalid area image processing unit 350 identifies a target area in the image data 501 based on position information notified by the target area detection unit 340. Image data 502 shows the state in which the target area has been identified. Also as shown in Fig. 5, the invalid area image processing unit 350 invalidates areas other than the target area and generates first invalidated image data 503.

[0057] 5, decoded data 511 is decoded data notified by the decoding unit 320. As shown in FIG. 5, the invalid area image processing unit 350 identifies the target area in the decoded data 511 based on the position information notified by the target area detection unit 340. Decoded data 512 shows the state in which the target area has been identified. As also shown in FIG. 5, the invalid area image processing unit 350 invalidates areas other than the target area and generates second invalidated image data 513.

[0058] Furthermore, as shown in FIG. 5, the invalid area image processing unit 350 calculates the difference between the first invalidated image data 503 and the second invalidated image data 513 to generate difference data 520.

[0059] <Hierarchical coding process flow> Next, a description will be given of the flow of the hierarchical coding process by the hierarchical coding section 121. Fig. 6 is a first flowchart showing the flow of the hierarchical coding process.

[0060] In step S601, the hierarchical encoding unit 121 acquires video data from the imaging device 110 on a frame-by-frame basis.

[0061] In step S602, the hierarchical encoding unit 121 determines the limit compression rate for one frame of image data obtained from the video data.

[0062] In step S603, the hierarchical coding unit 121 detects a target area required to recognize the recognition target in one frame of image data acquired from the video data, and outputs position information by calculating the position of the detected target area within the image data.

[0063] In step S604, the hierarchical encoding unit 121 encodes the entire area of ​​one frame of image data acquired from the video data at the limit compression rate to generate first encoded data.

[0064] In step S605, the hierarchical encoding unit 121 decodes the generated first encoded data to generate decoded data.

[0065] In step S606, the hierarchical encoding unit 121 generates first invalidated image data by invalidating an area other than the target area specified by the position information from one frame of image data acquired from the video data.

[0066] In step S607, the hierarchical encoding unit 121 generates second invalidated image data by invalidating areas of the generated decoded data other than the target area specified by the position information.

[0067] In step S608, the hierarchical encoding unit 121 calculates the difference between the first invalidated image data and the second invalidated image data, and generates difference data.

[0068] In step S609, the hierarchical encoding unit 121 encodes the generated differential data at a predetermined compression rate to generate second encoded data.

[0069] In step S610, the hierarchical encoding unit 121 transmits the generated first encoded data and the generated second encoded data to the server device .

[0070] In step S611, the hierarchical coding unit 121 determines whether or not to end the hierarchical coding process. If it is determined in step S611 that the hierarchical coding process is not to be ended (continued) (NO in step S611), the process returns to step S601.

[0071] On the other hand, if it is determined in step S611 that the hierarchical encoding process is to be ended (YES in step S611), the hierarchical encoding process is ended.

[0072] <Functional configuration of the hierarchical decoding unit of the server device> Next, a description will be given of the functional configuration of the hierarchical decoding unit 131 of the server device 130. Fig. 7 is a first diagram showing an example of the functional configuration of the hierarchical decoding unit of the server device. As shown in Fig. 7, the hierarchical decoding unit 131 has a decoding unit 710, a decoding unit 720, an addition unit 730, and a storage control unit 740.

[0073] The decoding unit 710 is an example of a first decoding unit. The decoding unit 710 receives first coded data transmitted from the image processing device 120. The decoding unit 710 also decodes the first coded data and notifies the adding unit 730 of the first decoded data.

[0074] The decoding unit 720 is an example of a second decoding unit. The decoding unit 720 receives second coded data transmitted from the image processing device 120. The decoding unit 720 also decodes the second coded data and notifies the adding unit 730 of the second decoded data.

[0075] The adding unit 730 adds the second decoded data notified by the decoding unit 720 to the first decoded data notified by the decoding unit 720, and generates decoded data after the addition.

[0076] The storage control unit 740 stores the generated decoded data after addition in the decoded data storage unit 132.

[0077] In this way, by compensating for the degradation of image quality caused by encoding at the limit compression rate for the target region, the hierarchical decoding unit 131 can generate decoded data for the target region that reproduces the image quality of the image data before encoding. Also, by decoding the first encoded data that was encoded at the limit compression rate without applying processing such as blacking out for regions other than the target region, the hierarchical decoding unit 131 can generate decoded data for regions other than the target region with improved image quality compared to conventional methods.

[0078] As a result, the decoded data can be used for purposes other than AI recognition processing. In other words, the hierarchical decoding unit 131 can generate highly convenient decoded data even when it is not possible to set different compression rates for each region.

[0079] <Specific example of the addition section processing> Next, a specific example of the processing of the addition unit 730 will be described. Fig. 8 is a diagram showing a specific example of the processing of the addition unit. In Fig. 8, first decoded data 801 is decoded data generated by the decoding unit 710 decoding the first encoded data. Also, second decoded data 802 is decoded data generated by the decoding unit 720 decoding the second encoded data.

[0080] As shown in FIG. 8, the adder 730 adds the second decoded data 802 to the first decoded data 801 to generate decoded data 810 after addition.

[0081] <Flow of layered decoding process> Next, a description will be given of the flow of hierarchical decoding processing by hierarchical decoding section 131. Fig. 9 is a first flowchart showing the flow of hierarchical decoding processing.

[0082] In step S901 , the hierarchical decoding unit 131 acquires the first encoded data transmitted from the image processing device 120 .

[0083] In step S902, the hierarchical decoding unit 131 decodes the acquired first encoded data to generate first decoded data.

[0084] In step S903 , the hierarchical decoding unit 131 acquires the second encoded data transmitted from the image processing device 120 .

[0085] In step S904, the hierarchical decoding unit 131 decodes the acquired second encoded data to generate second decoded data.

[0086] In step S905, the second decoded data is added to the first decoded data.

[0087] In step S906, the hierarchical decoding unit 131 stores the decoded data after the addition in the decoded data storage unit 132.

[0088] In step S907, the hierarchical decoding unit 131 determines whether or not to end the hierarchical decoding process. If it is determined in step S907 that the hierarchical decoding process is not to be ended (continued) (NO in step S907), the process returns to step S901.

[0089] On the other hand, if it is determined in step S907 that the hierarchical decoding process is to be ended (YES in step S907), the hierarchical decoding process is ended.

[0090] As is clear from the above description, the image processing device 120 according to the first embodiment determines a target area required to recognize a recognition target in image data and a limit compression rate at which the recognition target can be recognized. The image processing device 120 according to the first embodiment encodes the entire area of ​​the image data at the determined limit compression rate and transmits first encoded data. The image processing device 120 according to the first embodiment also generates first invalidated image data by invalidating areas of the image data other than the target area. The image processing device 120 according to the first embodiment also generates second invalidated image data by invalidating areas of the decoded data obtained by encoding the image data at the limit compression rate and then decoding the image data. The image processing device 120 according to the first embodiment also encodes differential data between the first invalidated image data and the second invalidated image data at a predetermined compression rate and transmits second encoded data.

[0091] As a result, according to the first embodiment, the image processing device 120 can reproduce the image quality of the image data before encoding for the target area, and can improve the image quality of areas other than the target area compared to conventional methods.

[0092] In addition, the server device 130 according to the first embodiment generates first decoded data by decoding the first encoded data, and adds second decoded data generated by decoding the second encoded data to the generated first decoded data.

[0093] As a result, according to the first embodiment, the server device 130 can generate decoded data for the target area that reproduces the image quality of the image data before encoding, and can generate decoded data for areas other than the target area that has improved image quality compared to conventional data.

[0094] As a result, according to the first embodiment, it becomes possible to use the decoded data for purposes other than AI recognition processing. In other words, according to the first embodiment, even if it is not possible to set different compression rates for each region, it becomes possible to generate highly convenient decoded data.

[0095] [Second embodiment] In the first embodiment described above, a target area detection unit was provided, and the target area detection unit detected the area necessary to recognize the recognition target in the image data, thereby outputting position information. However, the function of outputting position information is not limited to being realized by the target area detection unit.

[0096] For example, when the compression rate determination unit determines the limit compression rate, the position information may be output by detecting the target region from the decoded data. The second embodiment will be described below, focusing on the differences from the first embodiment.

[0097] <Functional configuration of the hierarchical coding unit of the image processing device> First, the functional configuration of the hierarchical encoding unit 121 of the image processing device 120 according to the second embodiment will be described. Fig. 10 is a third diagram showing an example of the functional configuration of the hierarchical encoding unit of the image processing device. The differences from the functional configuration described using Fig. 3 are that the target region detection unit 340 is not included and that the function of the compression rate determination unit 1010 is different from the function of the compression rate determination unit 330.

[0098] The compression rate determination unit 1010 performs AI recognition processing on the decoded data notified by the decoding unit 320, and determines whether or not the recognition target included in the decoded data can be recognized. If the compression rate determination unit 1010 determines that the recognition target can be recognized, it increases the compression rate by a predetermined increment and notifies the encoding unit 310.

[0099] Furthermore, if the compression rate determination unit 1010 determines that the recognition target cannot be recognized, it determines the compression rate previously notified to the encoding unit 310 as the limit compression rate. Furthermore, the compression rate determination unit 1010 notifies each of the encoding unit 310 and the encoding unit 360 of the determined limit compression rate.

[0100] The compression rate determination unit 1010 also detects, as a target region, a region necessary for recognizing a recognition target in decoded data when encoded at the limit compression rate. The compression rate determination unit 1010 also calculates the position of the detected target region in the decoded data and notifies the invalid region image processing unit 350 of the position information.

[0101] <Hierarchical coding process flow> Next, the flow of hierarchical coding processing by the hierarchical coding unit 121 of the image processing device 120 according to the second embodiment will be described. Fig. 11 is a second flowchart showing the flow of hierarchical coding processing. The difference from Fig. 6 is step S1101.

[0102] In step S1101, the hierarchical coding unit 121 detects a target area necessary for recognizing a recognition target in the decoded data when encoded at the limit compression rate, and outputs position information by calculating the position of the detected target area in the decoded data.

[0103] As is clear from the above description, the image processing device 120 according to the second embodiment detects, as a target area, an area required for recognizing a recognition target in decoded data when determining a limit compression rate, and outputs position information. As a result, according to the second embodiment, it is possible to obtain the same effects as those of the first embodiment.

[0104] [Third embodiment] In the first and second embodiments, the deterioration of image quality in the target area is compensated for, and the area other than the target area is encoded at the limit compression rate without applying processing such as blacking out, thereby improving the usability of the decoded data.

[0105] On the other hand, in the first and second embodiments, in order to compensate for the deterioration of image quality in the target region, difference data is generated from image data and coded.

[0106] In contrast to this, in the third embodiment, instead of compensating for the deterioration of image quality in the target region, the following is extracted from the image data: Image data in which areas other than the target area are invalidated (area image data for the enhance layer), Image data with the target area disabled (area image data for base layer), is generated and encoded. As a result, according to the third embodiment, as with the first and second embodiments, it is possible to improve the image quality of areas other than the target area compared to when image data that has been subjected to processing such as blacking out is encoded. In addition, according to the third embodiment, it is possible to reduce the amount of calculation required to decode the first encoded data compared to the first and second embodiments.

[0107] The third embodiment will be described below, focusing on the differences from the first and second embodiments.

[0108] <Functional configuration of the hierarchical coding unit of the image processing device> First, the functional configuration of a hierarchical encoding unit 121 of an image processing device 120 according to the third embodiment will be described with reference to Fig. 12 to Fig. 14. Fig. 12 to Fig. 14 are fourth to sixth diagrams showing an example of the functional configuration of the hierarchical encoding unit of an image processing device.

[0109] Here, the functions of each unit when encoding one frame of image data out of the image data of each frame included in video data will be explained in Figures 12, 13, and 14 according to the execution order.

[0110] As shown in Figures 12 to 14, the hierarchical coding unit 121 of the image processing device 120 according to the third embodiment has an encoding unit 310, a decoding unit 320, a compression rate determination unit 330, a target area detection unit 340, an invalid area image processing unit 1210, and an encoding unit 360.

[0111] First, the function of each unit up to the determination of the limit compression rate used when encoding one frame of image data will be described with reference to FIG.

[0112] The encoding unit 310 acquires one frame of image data from the video data, encodes the acquired image data using the compression rate notified by the compression rate determination unit 330, and notifies the decoding unit 320 of the encoded data.

[0113] The decoding unit 320 decodes the coded data notified by the encoding unit 310 to generate decoded data, and notifies the compression rate determination unit 330 of the generated decoded data.

[0114] The compression rate determination unit 330 performs AI recognition processing on the decoded data notified by the decoding unit 320, and determines whether or not the recognition target included in the image data can be recognized. If the compression rate determination unit 330 determines that the recognition target can be recognized, it increases the compression rate by a predetermined increment and notifies the encoding unit 310.

[0115] If the compression rate determination unit 330 determines that the recognition target cannot be recognized, it determines the compression rate previously notified to the encoding unit 310 as the limit compression rate. Furthermore, the compression rate determination unit 330 notifies the encoding unit 310 and the encoding unit 360 of the determined limit compression rate.

[0116] Next, the functions of each unit up to the generation of image data to be coded in the encoding unit 310 and the encoding unit 360 will be described with reference to FIG.

[0117] The target area detection unit 340 performs AI recognition processing on one frame of image data acquired from the video data, and detects the target area in the image data that is necessary to recognize the recognition target. The target area detection unit 340 also calculates the position of the detected target area within the image data, and notifies the invalid area image processing unit 1210 of the position information.

[0118] The invalid area image processing unit 1210 generates area image data for the enhance layer (second invalid image data) by invalidating areas other than the target area specified by the notified position information from image data for one frame acquired from the video data. Furthermore, the invalid area image processing unit 1210 notifies the encoding unit 360 of the generated area image data for the enhance layer (second invalid image data) together with the position information.

[0119] Furthermore, the invalid area image processing unit 1210 generates area image data for the base layer (first invalidated image data) by invalidating a target area identified by the notified position information from image data for one frame acquired from the video data. Furthermore, the invalid area image processing unit 1210 notifies the encoding unit 310 of the generated area image data for the base layer (first invalidated image data) together with the position information.

[0120] Next, the function of each unit up to the generation of the first coded data and the second coded data will be described with reference to FIG.

[0121] The encoding unit 310 generates first encoded data by encoding the area image data for the base layer (first invalid image data) notified by the invalid area image processing unit 1210 at a compression rate higher than the limit compression rate, and transmits the first encoded data to the server device 130 together with location information.

[0122] The encoding unit 310 generates second encoded data by encoding the area image data for the enhanced layer (second invalid image data) notified by the invalid area image processing unit 1210 at the limit compression rate, and transmits the second encoded data to the server device 130 together with the location information.

[0123] In this way, the hierarchical coding unit 121 disables areas other than the target area and then encodes the image data at the limit compression rate, and also disables the target area and then encodes the image data at a compression rate higher than the limit compression rate. As a result, according to the third embodiment, as with the first and second embodiments, it is possible to improve the image quality of areas other than the target area compared to when image data that has been processed, such as blacked out, is encoded. In addition, according to the third embodiment, it is possible to reduce the amount of calculation required to decode the first encoded data compared to the first and second embodiments.

[0124] Furthermore, in the case of the third embodiment, there is an advantage that when encoding a target region, there is no need to change the compression rate even if the resolution parameters of the encoding unit 310 or the number of target regions are increased or decreased to correspond to the resolution of the video data. Similarly, in the case of the third embodiment, there is an advantage that when encoding a region other than the target region, there is no need to change the compression rate even if the resolution parameters of the encoding unit 310 or the number of target regions are increased or decreased to correspond to the resolution of the video data.

[0125] <Specific example of processing by the invalid area image processing unit> Next, a description will be given of a specific example of the processing by the invalid area image processing unit 1210. Fig. 15 is a second diagram showing a specific example of the processing by the invalid area image processing unit.

[0126] 15, image data 1501 is image data for one frame acquired from video data. As shown in Fig. 15, the invalid area image processing unit 1210 identifies a target area in the image data 1501 based on position information notified by the target area detection unit 340. Image data 1502 shows the state in which the target area has been identified.

[0127] Also, as shown in FIG. 15, the invalid area image processing unit 1210 generates area image data 1503 for the enhance layer (second invalid image data) by invalidating areas other than the target area, and notifies the encoding unit 360 of this together with position information.

[0128] Similarly, the invalid area image processing unit 1210 invalidates the target area to generate area image data 1504 for the base layer (first invalidated image data), and notifies the encoding unit 310 of this together with the position information.

[0129] <Hierarchical coding process flow> Next, the flow of hierarchical coding processing by the hierarchical coding unit 121 of the image processing device 120 according to the third embodiment will be described. Fig. 16 is a third flowchart showing the flow of hierarchical coding processing. The differences from Fig. 6 are steps S1601 to S1605.

[0130] In step S1601, the hierarchical coding unit 121 identifies a target area based on position information in one frame of image data obtained from video data, and generates area image data for an enhanced layer by disabling areas other than the identified target area.

[0131] In step S1602, the hierarchical encoding unit 121 identifies a target area based on position information in one frame of image data obtained from video data, and generates area image data for the base layer by invalidating the identified target area.

[0132] In step S1603, the hierarchical encoding unit 121 encodes the region image data for the enhance layer at the limit compression rate to generate second encoded data.

[0133] In step S1604, the hierarchical encoding unit 121 encodes the region image data for the base layer at a compression rate higher than the limit compression rate to generate first encoded data.

[0134] In step S1605, the hierarchical encoding unit 121 transmits the first encoded data and the second encoded data to the server device 130 in association with the position information.

[0135] <Functional configuration of the hierarchical decoding unit of the server device> Next, the functional configuration of the hierarchical decoding unit 131 of the server device 130 according to the third embodiment will be described. Fig. 17 is a second diagram showing an example of the functional configuration of the hierarchical decoding unit of the server device. As shown in Fig. 17, the hierarchical decoding unit 131 includes a decoding unit 710, a decoding unit 720, a substitution unit 1710, and a storage control unit 740.

[0136] The decoding unit 710 receives the first coded data together with the position information transmitted from the image processing device 120. The decoding unit 710 also decodes the first coded data and notifies the replacing unit 1710 of the first decoded data together with the position information.

[0137] The decoding unit 720 receives the second coded data together with the position information transmitted from the image processing device 120. The decoding unit 720 also decodes the second coded data and notifies the replacement unit 1710 of the second decoded data together with the position information.

[0138] The replacement unit 1710 replaces the target area identified by the position information in the first decoded data notified by the decoding unit 710 with the target area identified by the position information in the second decoded data notified by the decoding unit 720, and generates the decoded data after replacement.

[0139] The storage control unit 740 stores the generated decoded data after substitution in the decoded data storage unit 132.

[0140] <Specific example of replacement processing> Next, a specific example of the processing by the permutation unit 1710 will be described. Fig. 18 is a diagram showing a specific example of the processing by the permutation unit. In Fig. 18, first decoded data 1801 is decoded data generated by the decoding unit 710 decoding the first encoded data. Also, second decoded data 1802 is decoded data generated by the decoding unit 720 decoding the second encoded data.

[0141] As shown in FIG. 18, the replacement unit 1710 replaces the target area identified by the position information in the first decoded data 1801 with the target area identified by the position information in the second decoded data 1802, thereby generating replaced decoded data 1810.

[0142] <Flow of layered decoding process> Next, the flow of hierarchical decoding processing by the hierarchical decoding unit 131 of the server device 130 according to the third embodiment will be described. Fig. 19 is a second flowchart showing the flow of hierarchical decoding processing. The differences from the first flowchart described using Fig. 9 are steps S1901 and S1902.

[0143] In step S1901, the hierarchical decoding unit 131 replaces the target area identified by the position information in the first decoded data obtained by decoding the first encoded data with the target area identified by the position information in the second decoded data obtained by decoding the second encoded data.

[0144] In step S1902, the hierarchical decoding unit 131 stores the decoded data after the substitution in the decoded data storage unit 132.

[0145] As is clear from the above description, the image processing device 120 according to the third embodiment encodes the second invalidated image data in which areas other than the target area are invalidated at the limit compression rate. Also, the image processing device 120 according to the third embodiment encodes the first invalidated image data in which the target area is invalidated at a compression rate higher than the limit compression rate.

[0146] As a result, according to the third embodiment, the image processing device 120 can improve the image quality of the area other than the target area more than ever before. In addition, according to the third embodiment, the amount of calculation required to decode the first encoded data can be reduced.

[0147] In other words, according to the third embodiment, even if it is not possible to set different compression rates for each region, it is possible to generate highly convenient decoded data and reduce the amount of calculation required for decoding.

[0148] [Fourth embodiment] In the above third embodiment, a target area detection unit is provided, and the target area detection unit detects the area necessary to recognize the recognition target in the image data, thereby outputting position information. However, the function of outputting position information is not limited to being realized by the target area detection unit.

[0149] For example, when the compression rate determination unit determines the limit compression rate, the position information may be output by detecting the target region from the decoded data. The following describes the fourth embodiment, focusing on the differences from the third embodiment.

[0150] <Functional configuration of the hierarchical coding unit of the image processing device> First, the functional configuration of the hierarchical encoding unit 121 of the image processing device 120 according to the fourth embodiment will be described with reference to Fig. 20 and Fig. 21. Fig. 20 and Fig. 21 are seventh and eighth diagrams showing an example of the functional configuration of the hierarchical encoding unit of the image processing device. The differences from the functional configuration described with reference to Fig. 12 are that the target region detection unit 340 is not included and that the function of the compression rate determination unit 2010 is different from the function of the compression rate determination unit 330.

[0151] First, with reference to FIG. 20, the functions of each unit from determining the limit compression rate used when encoding one frame of image data to outputting the position information will be described.

[0152] The compression rate determination unit 2010 performs AI recognition processing on the decoded data notified by the decoding unit 320, and determines whether or not the recognition target included in the decoded data can be recognized. If the compression rate determination unit 2010 determines that the recognition target can be recognized, it increases the compression rate by a predetermined increment and notifies the encoding unit 310.

[0153] Furthermore, if the compression rate determination unit 2010 determines that the recognition target cannot be recognized, it determines the compression rate previously notified to the encoding unit 310 as the limit compression rate. Furthermore, the compression rate determination unit 2010 notifies the encoding unit 310 and the encoding unit 360 of the determined limit compression rate.

[0154] The compression rate determination unit 2010 also detects, as a target region, a region necessary for recognizing a recognition target in the decoded data when encoded at the limit compression rate. The compression rate determination unit 2010 also calculates the position of the detected target region in the decoded data and notifies the invalid region image processing unit 1210 of the position information.

[0155] Next, the functions of each unit up to the generation of image data to be coded in the encoding unit 310 and the encoding unit 360 will be described with reference to FIG.

[0156] The invalid area image processing unit 1210 generates area image data for the enhance layer (second invalid image data) by invalidating areas other than the target area specified by the notified position information from image data for one frame acquired from the video data. Furthermore, the invalid area image processing unit 1210 notifies the encoding unit 360 of the generated area image data for the enhance layer (second invalid image data) together with the position information.

[0157] Furthermore, the invalid area image processing unit 1210 generates area image data for the base layer (first invalidated image data) by invalidating a target area identified by the notified position information from image data for one frame acquired from the video data. Furthermore, the invalid area image processing unit 1210 notifies the encoding unit 310 of the generated area image data for the base layer (first invalidated image data) together with the position information.

[0158] <Hierarchical coding process flow> Next, the flow of hierarchical encoding processing by the hierarchical encoding unit 121 of the image processing device 120 according to the fourth embodiment will be described. Fig. 22 is a fourth flowchart showing the flow of hierarchical encoding processing. The difference from the third flowchart described using Fig. 16 is step S2201.

[0159] In step S2201, the hierarchical coding unit 121 detects a target area necessary for recognizing a recognition target in the decoded data when encoded at the limit compression rate, and outputs position information by calculating the position of the detected target area in the decoded data.

[0160] As is clear from the above description, the image processing device 120 according to the fourth embodiment detects, as a target area, an area required for recognizing a recognition target in decoded data when determining a limit compression rate, and outputs position information. As a result, according to the fourth embodiment, it is possible to obtain the same effects as those of the third embodiment.

[0161] [Other embodiments] In the first and second embodiments, the invalid area image processing unit is described as generating differential data by calculating the difference between the first invalidation image data and the second invalidation image data. However, the method of generating differential data is not limited to this. For example, differential data may be generated by calculating the difference between the first invalidation image data and the second invalidation image data and then adding an offset value. This makes it possible to make the differential data non-negative.

[0162] In the first and second embodiments, the hierarchical encoding unit 121 generates two-layer coded data by taking an example in which the image data contains one recognition target. However, the method of generating coded data is not limited to this. For example, if the image data contains n recognition targets and the critical compression rates of the respective recognition targets are different, n+1 layers of coded data are generated.

[0163] In the third and fourth embodiments, the encoding unit 310 transmits the first encoded data together with the position information, and the encoding unit 360 transmits the second encoded data together with the position information. However, the position information may be transmitted by either the encoding unit 310 or the encoding unit 360 together with the encoded data.

[0164] In the above third and fourth embodiments, the invalid area image processing unit has been described as invalidating a target area specified by position information when generating area image data for a base layer (first invalid image data). Also, the invalid area image processing unit has been described as invalidating an area other than the target area specified by position information when generating area image data for an enhance layer (second invalid image data).

[0165] However, the method of generating the first invalidated image data and the second invalidated image data is not limited to this. For example, the first invalidated image data may be generated by invalidating an area smaller than the target area specified by the position information. Alternatively, the second invalidated image data may be generated by invalidating an area other than an area larger than the target area specified by the position information. In other words, when the replacement unit performs replacement, a margin area may be provided at the boundary portion of the target area so that the first decoded data and the second decoded data overlap. Note that, for example, a filtering process may be performed on the overlapping margin area. In this case, the filtering process may include, for example, a process of averaging pixel values ​​of the first decoded data and pixel values ​​of the second decoded data, but is not limited to this process.

[0166] In the third and fourth embodiments, the replacement unit performs replacement using position information transmitted and received between the hierarchical encoding unit and the hierarchical decoding unit. However, processing equivalent to the replacement performed by the replacement unit may be performed without transmitting and receiving position information. Specifically, the hierarchical encoding unit generates first invalidation image data that zeroes pixel values ​​in a target region of the first decoded data, and generates second invalidation image data that zeroes pixel values ​​in a region other than the target region of the second decoded data. The hierarchical decoding unit searches for a position where, when each pixel value of the first decoded data is added to each pixel value of the second decoded data, neither the pixel values ​​of the region other than the target region of the first decoded data nor the pixel values ​​of the target region of the second decoded data change. This allows the hierarchical decoding unit to properly align the first decoded data and the second decoded data, thereby achieving processing equivalent to replacement.

[0167] Furthermore, in the above-described embodiments, the hierarchical decoding unit 131 of the server device 130 has been described as storing the decoded data after addition or the decoded data after substitution in the decoded data storage unit 132. However, the function of the hierarchical decoding unit 131 is not limited to this, and for example, the hierarchical decoding unit 131 may perform filtering or the like before storing the data in order to reduce the difference in image quality between the target region and a region other than the target region.

[0168] Furthermore, in the above-described embodiments, the image processing device 120 has been described as functioning as the hierarchical encoding unit 121, and the server device 130 as functioning as the hierarchical decoding unit 131. However, the image processing device 120 may have some of the functions included in the hierarchical decoding unit 131 in addition to the hierarchical encoding unit 121. Alternatively, the server device 130 may have some of the functions included in the hierarchical encoding unit 121 in addition to the hierarchical decoding unit 131.

[0169] Furthermore, in each of the above embodiments, the imaging device 110 and the image processing device 120 have been described as separate devices, but the imaging device 110 and the image processing device 120 may be integrated into one device. Alternatively, the imaging device 110 may have some of the functions included in the hierarchical encoding unit 121 of the image processing device 120.

[0170] Furthermore, the AI-based recognition processing described in each of the above embodiments may include, in addition to deep learning processing, analysis processing that obtains results based on analysis by a computer or the like.

[0171] In each of the above embodiments, the process used by the target area detection unit 340 to detect the target area necessary for recognizing the recognition target and the process used by the compression rate determination unit 330 to determine whether the recognition target can be recognized may be the same process or different processes.

[0172] In addition, in each of the above embodiments, the compression rate determination unit 330 determines the limit compression rate by increasing the compression rate in predetermined increments, but the method for determining the limit compression rate is not limited to this. For example, the compression rate determination unit 330 may determine the limit compression rate by analyzing the recognition state and recognition process by the AI.

[0173] In each of the above embodiments, the compression rate used by the object area detection unit 340 when detecting the object area necessary for recognizing the recognition target, and the limit compression rate determined by the compression rate determination unit 330 may be compression rates according to the intended use of the decoded data. For example, The limit compression rate determined by the compression rate determination unit 330 may be a compression rate at which the recognition target can be recognized, The compression rate used by the object region detection unit 340 when detecting the object region necessary for recognizing the recognition target may be a compression rate suitable for analyzing the recognition target in more detail or with higher accuracy.

[0174] Furthermore, the method of determining the compression rates when generating the first encoded data and the second encoded data described in each of the above embodiments is merely an example, and other determination methods may be used. For example, one or both of the compression rates when generating the first encoded data and the compression rate when generating the second encoded data may be determined based on the results of AI analysis, information on the analysis process, etc.

[0175] The present invention is not limited to the configurations described in the above embodiments, but may be combined with other elements, etc. These aspects can be changed without departing from the spirit of the present invention, and can be appropriately determined depending on the application form. [Explanation of symbols]

[0176] 100: Image processing system 110: Imaging device 120: Image processing device 121: Layered encoding unit 130: Server device 131: Hierarchical decoding unit 310: Encoding section 320: Decoder 330: Compression ratio determination unit 340: Target area detection unit 350: Invalid area image processing unit 360: Encoding section 503: First invalidation image data 513: Second invalidation image data 520: Differential data 710: Decoder 720: Decoder 730: Addition section 740: Storage control unit 801: First decrypted data 802: Second decrypted data 810: Decoded data after addition 1010: Compression ratio determination unit 1210: Invalid area image processing section 1503: Base layer area image data (first invalidation image data) 1504: Enhance layer area image data (second invalidation image data) 1801: First decrypted data 1802: Second decrypted data 1810: Decoded data after substitution 2010: Compression ratio determination section

Claims

1. A determination unit that determines a target area required to recognize a recognition target in image data and a limit compression rate at which the recognition target can be recognized based on the result of the recognition processing by AI; a first encoding unit that encodes the entire area of ​​the image data at the limit compression rate and transmits the encoded data; a generating unit that generates first invalidated image data obtained by invalidating an area other than a target area of ​​the image data, and second invalidated image data obtained by invalidating an area other than the target area of ​​decoded data obtained by encoding the image data at the limit compression rate and then decoding the image data; a second encoding unit that encodes differential data between the first invalidated image data and the second invalidated image data at a predetermined compression rate and transmits the encoded differential data; An encoding device having:

2. 2. The encoding device according to claim 1, wherein the determination unit determines the limit compression rate based on a result of a recognition process obtained by performing the AI ​​recognition process on decoded data generated by decoding the encoded image data each time the compression rate is increased by a predetermined increment.

3. The encoding device according to claim 1 , wherein the differential data is generated by calculating a difference between the first invalidated image data and the second invalidated image data, and then adding an offset value to the calculated difference to make the difference non-negative.

4. 2. A decoding device in communication with the encoding device of claim 1, comprising: a first decoding unit that decodes the image data encoded by the first encoding unit; a second decoding unit that decodes the differential data encoded by the second encoding unit; an adding unit that adds the decoded difference data to the decoded image data; A decoding device having:

5. The decoding device according to claim 4 , further comprising: filtering a boundary portion between the target region and a region other than the target region in the image data after addition using the decoded difference data.

6. A determination unit that determines a target area required to recognize a recognition target in image data and a limit compression rate at which the recognition target can be recognized based on the result of the recognition processing by AI; a first encoding unit that encodes first invalidated image data obtained by invalidating an area of ​​the image data that is smaller than the target area at a compression rate higher than the limit compression rate and transmits the encoded first invalidated image data; a second encoding unit that encodes second invalidated image data obtained by invalidating an area of ​​the image data other than an area larger than the target area at the limit compression rate and transmits the second invalidated image data; An encoding device having:

7. 7. The encoding device according to claim 6, wherein the determination unit determines the limit compression rate based on a result of a recognition process obtained by performing the AI ​​recognition process on decoded data generated by decoding the encoded image data each time the compression rate is increased by a predetermined increment.

8. 7. The encoding device according to claim 6, wherein at least one of the first encoded data generated by encoding the first disabled image data and the second encoded data generated by encoding the second disabled image data includes information indicating the position of the target area.

9. A decoding device in communication with the encoding device of claim 6, comprising: a first decoding unit that decodes the first invalidated image data encoded by the first encoding unit; a second decoding unit that decodes the second invalidated image data encoded by the second encoding unit; a replacement unit that replaces an area of ​​the decoded first invalidated image data that is larger than the target area with an area of ​​the decoded second invalidated image data that is larger than the target area, and performs a filtering process on the overlapping area; A decoding device having:

10. Based on the results of the AI ​​recognition process, a target area required to recognize the recognition target in the image data and a limit compression rate at which the recognition target can be recognized are determined; Encoding the entire area of ​​the image data at the limit compression rate and transmitting the encoded data; generating first invalidated image data in which an area other than the target area is invalidated from the image data, and second invalidated image data in which an area other than the target area is invalidated from decoded data obtained by encoding the image data at the limit compression rate and then decoding the image data, The differential data between the first invalidated image data and the second invalidated image data is encoded at a predetermined compression rate and transmitted. A coding method that allows a computer to carry out the processing.

11. decoding the image data and the differential data encoded by the encoding method according to claim 10; The decoded difference data is added to the decoded image data. A computer-implemented decryption method.

12. Based on the results of the AI ​​recognition process, a target area required to recognize the recognition target in the image data and a limit compression rate at which the recognition target can be recognized are determined; encoding first invalidated image data, which is obtained by invalidating an area of ​​the image data that is smaller than the target area, at a compression rate higher than the limit compression rate, and transmitting the encoded first invalidated image data; The second invalidated image data obtained by invalidating the area of ​​the image data other than the area larger than the target area is encoded at the limit compression rate and transmitted. A coding method that allows a computer to carry out the processing.

13. decoding the first invalidated image data and the second invalidated image data encoded by the encoding method according to claim 12; Replace an area of ​​the decoded first invalidation image data that is larger than the target area with an area of ​​the decoded second invalidation image data that is larger than the target area, and perform a filtering process on the overlapping area. A computer-implemented decryption method.

14. Based on the results of the AI ​​recognition process, a target area required to recognize the recognition target in the image data and a limit compression rate at which the recognition target can be recognized are determined; Encoding the entire area of ​​the image data at the limit compression rate and transmitting the encoded data; generating first invalidated image data in which an area other than the target area is invalidated from the image data, and second invalidated image data in which an area other than the target area is invalidated from decoded data obtained by encoding the image data at the limit compression rate and then decoding the image data, The differential data between the first invalidated image data and the second invalidated image data is encoded at a predetermined compression rate and transmitted. An encoded program that causes a computer to execute a process.

15. decoding the image data and the differential data encoded by executing the encoding program according to claim 14; The decoded difference data is added to the decoded image data. A decryption program that causes a computer to execute the process.

16. Based on the results of the AI ​​recognition process, a target area required to recognize the recognition target in the image data and a limit compression rate at which the recognition target can be recognized are determined; encoding first invalidated image data, which is obtained by invalidating an area of ​​the image data that is smaller than the target area, at a compression rate higher than the limit compression rate, and transmitting the encoded first invalidated image data; The second invalidated image data obtained by invalidating the area of ​​the image data other than the area larger than the target area is encoded at the limit compression rate and transmitted. An encoded program that causes a computer to execute a process.

17. decoding the first invalidated image data and the second invalidated image data that have been coded by executing the coding program according to claim 16; Replace an area of ​​the decoded first invalidation image data that is larger than the target area with an area of ​​the decoded second invalidation image data that is larger than the target area, and perform a filtering process on the overlapping area. A decryption program that causes a computer to execute the process.

Images