Real-time video data high-reliability transmission method and system

Abstract

The invention discloses a real-time video data high-reliability transmission method and system, and the method comprises the steps: adding a plurality of preprocessing steps before the coding of original video data, firstly dividing each frame of video frame data into a plurality of video blocks, and respectively embedding a completeness identifier for each video block, the method comprises the following steps: firstly, acquiring video frame data, then preprocessing the video frame data based on a pixel point displacement mechanism to form a plurality of preprocessed video frames, meanwhile, constructing a plurality of pieces of video component data based on the plurality of generated preprocessed video frames, normally carrying out operations such as video coding and message packaging, and then, adjusting a TOS field in a video message to form a plurality of pieces of video component data. According to the method, the multiple video component data are subjected to multi-channel parallel transmission, finally, the multiple video component data are combined at a receiving end based on the embedded integrity identifier, and the video data closest to the original video data are recovered. According to the invention, the reliability of video data transmission is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data transmission, and in particular to a high-reliability transmission method and system for real-time video data. Background Art

[0002] Since IP-based video services such as video calls, video conferencing, and video surveillance mainly rely on IP communication networks such as the Internet for carrying and transmission, and the Internet is a typical global network jointly built and operated by multiple parties, its transmission quality and network performance have obvious regional differences. The communication bandwidth, packet loss rate and other data transmission indicators in some regions may be significantly inferior to those in other regions. This poses a huge challenge to the reliable transmission of end-to-end video data. Once the video transmission path passes through a data exchange node with relatively poor transmission quality in the Internet, it may face obvious packet loss or bit errors, resulting in mosaic images or video signal interruption at the receiving end, seriously affecting the transmission effect of video services.

[0003] The existing technology mainly achieves the enhancement of the reliability of video data transmission by optimizing the video coding algorithm or adjusting the video coding parameters. It relies on the high-redundancy video coding algorithm to make the video data have a certain redundancy after coding, which can resist packet loss and bit errors during transmission to a certain extent. However, in the actual application process, the existing technology has two shortcomings: first, the most commonly used video coding algorithms are only H.264 and H.265, and the effects of different video coding algorithms in resisting transmission errors are similar. Switching the coding algorithm is difficult to bring about a qualitative improvement in the reliability of video transmission; second, the typical video coding algorithm and the legal coding parameter range often need to consider the balance between communication performance and video transmission reliability, and do not support the introduction of excessive data redundancy by adjusting the coding parameters. In the face of a harsh network transmission environment, optimizing the video coding parameters cannot effectively improve the reliability of video data transmission. Summary of the invention

[0004] In view of this, the present invention provides a high-reliability transmission method and system for real-time video data, which solves the problem that the prior art is heavily dependent on the error correction capability of the video encoding and decoding algorithm itself and is difficult to adapt to extremely harsh communication environments, thereby significantly improving the transmission reliability of real-time video data under any conditions.

[0005] The present invention discloses a real-time video data high-reliability transmission method, which divides original video data into multiple blocks, adds integrity identifiers to each block, pre-processes the blocks to generate multiple video component data containing the same amount of information, different video component data carry different network communication identifiers, transmits the video component data in a corresponding network path according to the communication identifier, and merges the data of multiple video components at the receiving end according to the video integrity identifier, so as to effectively improve the reliability of video data transmission, including the following steps:

[0006] Step 1, video block division: according to a pre-agreed method, each frame of video is divided into multiple different video blocks using a pixel point set representation method, each video block contains a different pixel point set, different video blocks do not contain the same pixel points, and all video blocks contain all the pixel points of the entire video frame;

[0007] Step 2: embedding integrity identifier: using an agreed method, embed a piece of integrity identifier data for each video block, which serves as a basis for the receiving end to judge the integrity of the block information;

[0008] Step 3, preprocessing of original video frame data: using a preprocessing method based on pixel displacement, one original video frame data is used to generate N preprocessed video frame data, each preprocessed video frame data contains the same amount of information as the original video frame data, where N represents the number of video components planned to be generated in the scheme;

[0009] Step 4, video component generation: generating N video component data based on the N pre-processed video frame data according to the pre-agreed redundancy and generation method;

[0010] Step 5, multi-channel parallel transmission control: by specifying different network communication identification fields for different video component data when encapsulating the video service message, the transit router is guided to forward different video components according to different forwarding strategies, so as to achieve the effect of multi-channel transmission of video component data;

[0011] Step 6, video component merging: The receiving end merges the received multiple video component data, and for the same video block in different video component data, selects the video block with the highest integrity to form the final video data according to the embedded data integrity identifier.

[0012] Furthermore, the step 1 comprises:

[0013] For the same frame of video data, the sending end of the video data and the receiving end of the video data use the same rule to realize the video block division. The rule is expressed in the form of a pixel point set, which is generated by a prior agreement or by the same algorithm and parameters. The method of using a pixel point set to represent the video block is as follows: for each frame of original video data, firstly, all the pixel points constituting the frame of video data are numbered according to a unified rule. For example, for video data with a resolution of m×n, there are m×n pixel points in a frame, which are numbered from 1 to m×n respectively. Then, according to the required video block integrity, the accuracy is identified and the appropriate block size is selected as needed. The block size should be able to divide the total number of pixel points. For example, for video data with a resolution of m×n, m is selected as the block size. Finally, a pixel point set is generated for each video block according to a pre-agreed method. For example, for video data with a resolution of m×n, a video block is formed by each row of pixel points, and each set includes m pixel points in total. Zi represents the i-th set, and the generated n sets are respectively

[0014] Z1{1,2,…,m},

[0015] Z2{m+1,m+2,…,m+m},

[0016] …,

[0017] Zn{m×(n-1)+1, m×(n-1)+2,…,m×n};

[0018] The set generation method is highly flexible. This solution supports custom generation of pixel point sets representing video blocks in any manner such as discontinuous numbering.

[0019] Furthermore, the step 2 comprises:

[0020] The RGB format data of each video block is processed with integrity identification embedding. The method for embedding integrity identification data for each video block is as follows: first, a random number function rand() with uniform distribution characteristics is used to generate a binary sequence M = {m1, m2, m3, ...} with a length of 3 times the total number of pixels in the video block frame. The value of each element in the sequence is 0 or 1, and the sequence must be consistent at both ends of the video data receiving and sending; then, each element in the binary sequence M is used in turn to replace the lowest bit of the R, G, and B data of each pixel in the video block frame, that is, each pixel is embedded with 3 bits of information from the binary sequence M; finally, different binary sequences can be generated for different video block frames, and the corresponding relationship between multiple binary sequences and video blocks must be consistent at both ends of the video data receiving and sending.

[0021] Furthermore, the step 3 comprises:

[0022] The method for preprocessing the original video frame data based on pixel point displacement is as follows: first, for the original video frame data with a resolution of m×n, number them according to the pixel point numbering rule to form a pixel point coding sequence S0={1, 2, 3, ..., m×n}; then, based on S0, randomly disrupt the order of pixels in S0 to form a new pixel point sequence S1. If the pixels in the sequence are backfilled into the original video frame according to the above pixel point numbering rule, a new video frame data that is completely different from the original video frame data will be formed, and the new video frame data is used as one of the preprocessed video frame data; finally, repeat the above steps to generate N-1 preprocessed video frame data, plus the original video frame data, to form a total of N preprocessed video frame data, where N represents the number of video components planned to be generated in the scheme.

[0023] Furthermore, the step 4 comprises:

[0024] The method for generating multiple video components is as follows: first, determine a video component redundancy k, the value range is [0,1], 0 represents zero redundancy, 1 represents 100% redundancy; then, based on the redundancy k, generate a random binary sequence B = {B1, B2, B3, ..., B m×n}, the proportion of "1" in the binary sequence is equal to the value of k, where m×n represents the resolution of the video frame to be processed; again, the random binary sequence B is used as the video frame mask, and is "ANDed" with the preprocessed video frame data generated by pixel point displacement in the above step, and all elements in the sequence B are traversed one by one. When the element is "1", the corresponding video pixel point data in the preprocessed video frame data remains unchanged, and when the element is "0", the corresponding video pixel point data in the preprocessed video frame data is reset to "all zero value"; finally, repeat the above steps N-1 times until all video frame data preprocessed based on pixel point displacement are processed, that is, N-1 preprocessed video frame data and N-1 different binary sequences B are used to form N-1 new video frame data, which together with the initial original video frame data constitute N video component data.

[0025] Furthermore, the step 5 comprises:

[0026] The method for realizing multi-channel parallel transmission is as follows: first, the generated N video component data are encoded as N independent video data streams to generate N video encoded video data, and IP layer data encapsulation is performed in a conventional manner; then, when filling the TOS field of the video data IP message, five types of general service (0000), maximum throughput (0100), highest reliability (0010), minimum delay (1000), and minimum cost (0001) are filled in a loop respectively, and the transit router is guided to forward different video component data according to different path forwarding strategies, so as to achieve the effect of multi-channel transmission of video component data.

[0027] Furthermore, the step 6 comprises:

[0028] The method for the receiving end to realize the merging of multiple video components is as follows: after the video data is decoded, all video blocks in each video component data are traversed, the integrity identifiers of the same video blocks in different video component data are compared one by one, and the video block with the highest integrity score is selected as the final received video block; after all the final received video blocks are selected, they will form a complete frame of video frame data.

[0029] Furthermore, in step 6, the method for calculating the completeness score is: after extracting the completeness identification sequence Q = {Q1, Q2, Q3, ...} enclosed in the video block, compare it with the original completeness identification M = {m1, m2, m3, ...} embedded by the sending end, and calculate the ratio P% of the number of set elements with consistent values ​​to the total number of elements, then P is the completeness score of the video block.

[0030] The present invention discloses a real-time video data high-reliability transmission system, which implements any of the above-mentioned real-time video data high-reliability transmission methods, and comprises:

[0031] Video block division module: used to divide each frame of input video data into video blocks to determine the minimum data unit for reliable transmission control of the system;

[0032] Integrity identification embedding module: used to embed a video data integrity identification into each video block as a basis for judging the integrity of the video block after transmission;

[0033] Video frame preprocessing module: used to expand one frame of original video frame data into multiple frames of preprocessed video frame data containing the same information by pixel displacement;

[0034] Video component generation module: used to cut the pre-processed video frame data containing multiple frames of the same information into video component frame data with moderate redundancy according to reliability requirements and channel packet loss conditions;

[0035] Multi-channel parallel transmission module: used to control the transmission parameters of the encoded video message to achieve the effect of multi-path video data parallel transmission;

[0036] Video component merging module: used to merge multiple received video component data according to the video block integrity information, and restore the video data that is closest to the original video frame.

[0037] Due to the adoption of the above technical solution, the present invention has the following advantages:

[0038] 1. The present invention changes the transmission method of video data between the transmitting end and the receiving end. By adding a multi-channel parallel transmission control step, the same original video data is split into multiple video component data containing incremental information and repeated information, which are distributed to multiple different network paths for transmission. The receiving end then merges the multiple video component data to restore the most complete original video data. This transmission method overcomes the defect of poor single-path transmission quality when based on single-path transmission, and significantly optimizes the end-to-end transmission reliability of video data.

[0039] 2. The present invention significantly improves the ability to resist sudden full-channel quality fluctuations under multi-channel parallel transmission conditions. By adding a video frame preprocessing step based on pixel displacement before generating multiple video components, the position of the same pixel in a single-frame video image belonging to different video components is significantly changed, thereby changing the position of the same pixel in the video data stream belonging to different video components, so that when multiple video components are synchronously transmitted through multiple channels in the network, when there are large-scale instantaneous transmission quality fluctuations that affect all transmission paths, such as rapid fading of the wireless network and congestion of the backbone path, the affected pixels in different video components are not the same pixels in the original video data, so that the original video data obtained after merging different video components under multi-channel transmission conditions will not be affected by the instantaneous transmission quality fluctuations of the entire path, and the reliability of video data transmission under multi-channel transmission conditions is significantly improved.

[0040] 3. The present invention proposes a new method for high-quality merging of multiple video components based on video integrity identification. By subdividing the original video frame data into multiple video data blocks, and modifying the lowest bit value of each pixel R, G, B data in the video block to the preset integrity identification data, and finally realizing the video block selection and combination mode among multiple video components based on the video block integrity score at the receiving end, on the one hand, it realizes the carrying of video data integrity information without increasing the additional network load and not significantly affecting the video quality, and effectively balances the contradiction between the video transmission quality and the network communication bandwidth. On the other hand, it provides a highly fine-grained multi-video component data merging mode, which is no longer limited to the coarse-grained optimization based on video frames or video streams, but is directly based on the optimization of pixel point (block) granularity. When multiple video components suffer from different degrees of quality damage at the same time (for example, blocks 1, 2, 3 of component 1 are damaged, and blocks 4, 5, 6 of component 2 are damaged), the video quality of the original video data generated after the merging of multiple video components can be significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0042] Figure 1 A structural schematic diagram of a real-time video data high-reliability transmission system according to an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a process for implementing video block division according to an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of a process for implementing embedding of a video block integrity identifier according to an embodiment of the present invention;

[0045] Figure 4 A schematic diagram of a process for implementing video frame preprocessing based on pixel point displacement according to an embodiment of the present invention;

[0046] Figure 5 A schematic diagram of a flow chart for realizing generation of multiple video components according to an embodiment of the present invention;

[0047] Figure 6 A schematic diagram of a flow chart for realizing multi-channel parallel transmission control of different video components according to an embodiment of the present invention;

[0048] Figure 7 It is a schematic diagram of a process of merging different video components based on video block integrity scoring according to an embodiment of the present invention;

[0049] Figure 8 A comparison diagram of average frame loss rate of video transmission using and not using an embodiment of the present invention;

[0050] Fig. 9 The figure is a comparison chart of average frame loss rates of video transmission using different numbers of video components in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The present invention is further described in conjunction with the accompanying drawings and embodiments, and the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field should fall within the scope of protection of the embodiments of the present invention.

[0052] like Figure 1 As shown, the present invention provides an embodiment of a method and system for highly reliable transmission of real-time video data, including a video block division module, an integrity identifier embedding module, a video frame preprocessing module, a video component generation module, a multi-channel parallel transmission module, and a video component merging module. The modules are interconnected through functional interfaces to jointly achieve highly reliable transmission of real-time video data.

[0053] After the video frame data to be sent is input into the system, it is first sent to the video block division module. According to the preset video block division method consistent with the receiving and sending ends, the input video frame data is divided into several video blocks. At the same time, after the integrity identifier embedding module receives the video frame data after the block division from the video block division module, it embeds an invisible integrity identifier for each video block respectively; then, the video frame embedded with the integrity identifier is sent to the video frame preprocessing module, and different video frame data after multiple pixel point displacements are generated according to the required number of video components, and the video component generation module generates multiple mutually redundant video component frame data based on the redundancy strategy; finally, after the multiple video component data are encoded and encapsulated, the multi-channel parallel transmission module marks different TOS field values ​​in the IP header to achieve the effect of multi-channel transmission of different video component data. After the receiving end receives different video component data transmitted through multiple paths, the video component merging module merges the video components according to the block integrity to achieve high-reliability transmission of video service data.

[0054] The processing of video data by the method of the present invention mainly includes six parts: video block division, integrity mark embedding, video frame preprocessing, video component generation, multi-channel parallel transmission, and video component merging.

[0055] like Figure 2As shown, the steps to implement video block division are: 1. After receiving the video frame, count the number of pixels m×n in the video frame in this session according to the video resolution m×n; 2. Calculate the divisor of m×n; 3. Starting from the smallest divisor, traverse all divisors and select a divisor of appropriate size as the video block size. The default selection is m (the number of pixels in a row); 4. According to the selected video block size, divide all pixels in the video frame into several video blocks. When the block size is m, the default continuous division method is used, that is, each row of pixels is a video block.

[0056] like Figure 3 As shown, the steps to embed the integrity identifier of the video block are as follows: 1. According to the selected block size, generate a binary random sequence M with a length equal to 3 times the total number of pixels in the block, and all elements in the sequence are 0 or 1; 2. Traverse all pixels in the block in turn, and replace the least sensitive 3 bits of data in each pixel. For the first pixel, the first element in M ​​replaces the lowest bit of the R component in the pixel, the second element replaces the lowest bit of the G component in the pixel, and the third element replaces the lowest bit of the B component in the pixel; 3. Bits 4-6 in M ​​are used to replace the second pixel data, bits 7-9 are used to replace the third pixel data, and so on, until all pixel data in the video block are replaced.

[0057] like Figure 4 As shown, the steps of implementing video frame preprocessing based on pixel displacement are: 1. Determine the required number of video components, which is 2 by default; 2. Use the original order of pixels to form a pixel coding sequence S0, S0 uses the same order as the pixel arrangement in the original video frame, that is, the video component frame generated according to S0 is essentially the original video frame; 3. Use the reverse order of pixels to form a pixel coding sequence S1, S1 uses the order completely opposite to the pixel arrangement in the original video frame, that is, the video component frame generated according to S1 is essentially the effect of the original video frame rotated 180 degrees; 4. According to the number of video components required, use other arbitrary strategies to generate pixel coding sequences, and generate preprocessed video frames accordingly.

[0058] like Figure 5 As shown in FIG. 1 , the steps for realizing the generation of multiple video components are as follows: 1. predefine a video component redundancy k, which is 0.7 by default; 2. generate a random binary sequence B = {B1, B2, B3, …, B m×n}, the proportion of "1" in the binary sequence is equal to the value of k, where m×n represents the resolution of the video frame to be processed; third, use the random binary sequence B as the video frame mask, and "AND" it with the preprocessed video frame data generated by pixel point displacement in the above steps, traverse all elements in the sequence B one by one, when the element is "1", the corresponding video pixel point data in the preprocessed video frame data remains unchanged, when the element is "0", the corresponding video pixel point data in the preprocessed video frame data is reset to "all zero value", and the generation of one frame of video frame data of one of the video components is completed; fourth, according to the number of video components N to be generated, a total of N-1 different binary sequences B are generated, and the "AND" step based on the binary sequence B is repeated N-1 times to complete the generation of N-1 video component frames, plus the initial original video frame data, together constitute N video component frame data.

[0059] like Figure 6 As shown, the steps for realizing multi-channel parallel transmission control of different video components are as follows: 1. If the video component belongs to the first video component, the TOS field of its IP message is assigned the general service identifier (0000); 2. If the video component belongs to the second video component, the TOS field of its IP message is assigned the maximum throughput identifier (0100); 3. If the video component belongs to the third video component, the TOS field of its IP message is assigned the highest reliability identifier (0010); 4. If the video component belongs to the fourth video component, the TOS field of its IP message is assigned the minimum delay identifier (1000); 5. If the video component belongs to the fifth video component, the TOS field of its IP message is assigned the minimum cost identifier (0001); 6. More video component data are cyclically selected and assigned in the five TOS field identifiers.

[0060] like Figure 7 As shown, the steps to achieve the merging of different video components are: 1. Traverse each video block in the video component frame in turn, extract the integrity identifier and calculate the video block integrity score; 2. In all video component frames, for the same video block, select the video block with the highest integrity score as the final preferred video block; 3. The pixel point set of all preferred video blocks constitutes the original video frame data after the multiple video components are merged.

[0061] The positive effects of the present invention are as follows: when the original video data is split into multiple video components and transmitted simultaneously through multiple different paths, it can significantly resist the quality fluctuation of a single transmission path and significantly improve the overall transmission reliability; at the same time, a preprocessing step based on pixel point displacement is added before the multiple video components are generated, so that the spatial / temporal distribution of the same pixel point in different video components changes significantly. When the quality of multiple transmission paths suddenly fluctuates at the same time, the pixels affected by different components are not the same, which significantly improves the video quality after the multiple components are merged; finally, in order to achieve the optimal video component merging effect, a fine-grained optimization and merging mechanism based on video blocks is adopted. When the blocks affected by different components are different, the original video data of the sender can be perfectly restored, which significantly improves the end-to-end transmission reliability of the video end data.

[0062] In order to verify the effect of improving the reliability of video transmission after adopting the present invention, simulation verification was carried out for the technology proposed by the present invention. Figure 8 As shown in the figure, when the embodiment of the present invention is not adopted, the average frame loss rate of video transmission increases significantly with the increase of channel packet loss rate, while after the embodiment of the present invention is adopted, the average frame loss rate of video transmission does not increase significantly with the increase of channel packet loss rate, and the video transmission frame loss rate is greatly reduced compared with the case when the embodiment of the present invention is not adopted, indicating that the present invention has a significant effect in reducing the frame loss rate of video transmission. Fig. 9 As shown, under different packet loss rate conditions, as the number of video components applied increases, the frame loss rate of video data is significantly reduced, indicating that the multi-component multi-channel transmission in the present invention has a significant resistance effect on the packet loss rate of the underlying channel. From the simulation results, it can be seen that the adoption of the present invention can significantly improve the reliability of video transmission in a harsh communication environment. The larger the number of video components, the higher the reliability. In actual use, the number of video components should also be determined by comprehensively considering factors such as channel bandwidth, transmission path, and terminal computing power.

[0063] The key points of the technical solution of the present invention are as follows:

[0064] Key point 1 of the technical solution: Added video component generation and multi-channel parallel transmission control steps: By splitting the same original video data into multiple redundant video component data and distributing them to multiple different network paths for transmission, the reduction in the transmission quality of a single video component does not affect the overall transmission quality of the original video data, thereby enhancing the ability to resist fluctuations in the transmission quality of a single network path during video transmission.

[0065] Key point 2 of the technical solution: Added a preprocessing step for the original video frame data: By adding a video frame preprocessing step based on pixel displacement before generating multiple video components, the position of the same pixel in a single-frame video image belonging to different video components is significantly changed, thereby changing the spatial / temporal distribution of the same pixel in the video data streams belonging to different video components. When multiple video components are transmitted synchronously through multiple channels in the network, when there are large-scale instantaneous transmission quality fluctuations that affect all transmission paths, such as rapid fading of the wireless network and congestion of the backbone path, the affected pixels in different video components are not the same pixels in the original video data. Therefore, the original video data obtained after merging different video components under multi-channel transmission conditions will not be affected by the instantaneous transmission quality fluctuations of the entire path, thereby enhancing the ability to resist simultaneous fluctuations in the transmission quality of multiple network paths during video transmission.

[0066] Key point 3 of the technical solution: Added video block division, integrity identification embedding, and video component merging steps: By subdividing the original video frame data into multiple video data blocks, and embedding integrity identification data for each video block, and finally implementing the video block optimization and combination method among multiple video components based on the video block integrity score at the receiving end, the quality of video data merging is enhanced when multiple video components suffer from different degrees of quality damage at the same time.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A high-reliability transmission method for real-time video data, characterized in that: include: (1) Decomposing the original video data into multiple video component data containing incremental information and repeated information, distributing them to different network paths for transmission, and merging the multiple video component data at the receiving end; (2) before decomposing the original video data into different video component data, a different video preprocessing step is added to change the position correlation between different pixels in the video frame; (3) The original video data is further subdivided into multiple video data blocks. As the minimum identification unit of the integrity of the original video data, an integrity marking step is added to each video data block before encoding. Without affecting the video picture, data integrity identification data is embedded as a basis for merging multiple video component data.

2. A method for high-reliability transmission of real-time video data according to claim 1, characterized in that: include: Step 1: Video block division: According to the pre-agreed method, each frame of video is divided into multiple different video blocks using the representation method of pixel point sets. Each video block contains a different set of pixel points. Different video blocks do not contain the same pixel points. All video blocks contain all the pixel points of the entire video frame. Step 2: Integrity identification embedding: Using an agreed method, a piece of integrity identification data is embedded in each video block, which serves as the basis for the receiving end to judge the integrity of the block information; Step 3: Preprocessing of original video frame data: Using a preprocessing method based on pixel displacement, one original video frame data is used to generate N preprocessed video frame data, each preprocessed video frame data contains the same amount of information as the original video frame data, where N represents the number of video components planned to be generated in the scheme; Step 4: Video component generation: Generate N video component data carrying incremental information based on the N pre-processed video frame data according to the pre-agreed redundancy and generation method; Step 5: Multi-channel parallel transmission control: By specifying different network communication identification fields for different video component data when encapsulating the video service message, the transit router is guided to forward different video components according to different forwarding strategies, so as to achieve the effect of multi-channel transmission of video component data; Step 6: Video component merging: The receiving end merges the received multiple video component data. For the same video block in different video component data, the video block with the highest integrity is selected according to the embedded data integrity identifier to form the final video data.

3. A method for high-reliability transmission of real-time video data according to claim 2, characterized in that: In step 1: For the same frame of video data, the sending end of the video data and the receiving end of the video data use the same rule to realize the video block division, and the rule is expressed in the form of a pixel point set, which is generated by prior agreement or by the same algorithm and parameters; the method of using the pixel point set to represent the video block is as follows: for each frame of original video data, firstly, all the pixel points constituting the frame of video data are numbered according to a unified rule, and for video data with a resolution of m×n, one frame has a total of m×n pixel points, which are numbered from 1 to m×n respectively; then, according to the required video block integrity, the accuracy is identified, and the appropriate block size is selected as needed, and the block size should be able to divide the total number of pixel points; finally, a pixel point set is generated for each video block according to a pre-agreed method, and the set contains all the pixel point numbers in the block; this scheme supports the use of any method to customize the generation of a pixel point set representing a video block; the arbitrary method includes discontinuous numbering.

4. A high-reliability transmission method for real-time video data according to claim 2, characterized in that: In step 2: The RGB format data of each video block is processed with integrity identification embedding. The method for embedding integrity identification data for each video block is as follows: first, a random number function rand() with uniform distribution characteristics is used to generate a binary sequence M = {m1, m2, m3, ...} with a length of 3 times the total number of pixels in the video block frame. The value of each element in the sequence is 0 or 1, and the sequence must be consistent at both ends of the video data receiving and sending; then, each element in the binary sequence M is used in turn to replace the lowest bit of the R, G, and B data of each pixel in the video block frame, that is, each pixel is embedded with 3 bits of information from the binary sequence M; finally, different binary sequences can be generated for different video block frames, and the corresponding relationship between multiple binary sequences and video blocks must be consistent at both ends of the video data receiving and sending.

5. A method for high-reliability transmission of real-time video data according to claim 2, characterized in that: In step 3: The method for preprocessing original video frame data based on pixel point displacement is as follows: first, for the original video frame data with a resolution of m×n, number them according to the pixel point numbering rule to form a pixel point coding sequence S0={1, 2, 3, ..., m×n}; then, based on S0, randomly disrupt the order of pixels in S0 to form a new pixel point sequence S1. If the pixels in the sequence are backfilled into the original video frame according to the pixel point numbering rule, a new video frame data completely different from the original video frame data will be formed, and the new video frame data is used as one of the preprocessed video frame data; finally, repeat the above steps to generate N-1 preprocessed video frame data, plus the original video frame data, to form a total of N preprocessed video frame data, where N represents the number of video components planned to be generated in the scheme.

6. A high-reliability transmission method for real-time video data according to claim 2, characterized in that: In step 4: The method for generating multiple video components is as follows: first, determine a video component redundancy k, the value range is [0,1], 0 represents zero redundancy, 1 represents 100% redundancy; then, based on the redundancy k, generate a random binary sequence B = {B1, B2, B3, ..., B m×n }, the proportion of "1" in the binary sequence is equal to the value of k, where m×n represents the resolution of the video frame to be processed; again, the random binary sequence B is used as the video frame mask, and is "ANDed" with the preprocessed video frame data generated by pixel point displacement in the above step, and all elements in the sequence B are traversed one by one. When the element is "1", the corresponding video pixel point data in the preprocessed video frame data remains unchanged, and when the element is "0", the corresponding video pixel point data in the preprocessed video frame data is reset to "all zero value"; finally, the above steps are repeated N-1 times until all video frame data preprocessed based on pixel point displacement are processed, that is, N-1 preprocessed video frame data and N-1 different binary sequences B are used to form N-1 new video frame data, which together with the initial original video frame data constitute N video component data.

7. A high-reliability transmission method for real-time video data according to claim 2, characterized in that: In step 5: The method for realizing multi-channel parallel transmission is as follows: first, the generated N video component data are encoded as N independent video data streams to generate N video encoded video data, and IP layer data encapsulation is performed in a conventional manner; Then, when filling the TOS field of the video data IP packet, five types of general service (0000), maximum throughput (0100), highest reliability (0010), minimum delay (1000), and minimum cost (0001) are filled in a loop respectively, guiding the transit router to forward different video component data according to different path forwarding strategies, so as to achieve the effect of multi-channel transmission of video component data.

8. A method for high-reliability transmission of real-time video data according to claim 2, characterized in that: In step 6: The method for the receiving end to realize the merging of multiple video components is as follows: after the video data is decoded, all video blocks in each video component data are traversed, the integrity identifiers of the same video blocks in different video component data are compared one by one, and the video block with the highest integrity score is selected as the final received video block; after all the final received video blocks are selected, they will form a complete frame of video frame data.

9. A method for high-reliability transmission of real-time video data according to claim 8, characterized in that: After extracting the integrity identifier sequence Q={Q1,Q2,Q3,…} enclosed in the video block, compare it with the original integrity identifier M={m1,m2,m3,…} embedded by the sender, and calculate the ratio P% of the number of set elements with consistent values ​​to the total number of elements, then P is the integrity score of the video block.

10. A real-time video data high-reliability transmission system, implementing the real-time video data high-reliability transmission method according to any one of claims 1 to 9, characterized in that: include: Video block division module: used to divide each frame of input video data into video blocks to determine the minimum data unit for reliable transmission control of the system; Integrity identification embedding module: used to embed a video data integrity identification into each video block as a basis for judging the integrity of the video block after transmission; Video frame preprocessing module: used to expand one frame of original video frame data into multiple frames of preprocessed video frame data containing the same information by pixel displacement; Video component generation module: used to cut the pre-processed video frame data containing multiple frames of the same information into video component frame data with moderate redundancy according to reliability requirements and channel packet loss conditions; Multi-channel parallel transmission module: used to control the transmission parameters of the encoded video message to achieve the effect of multi-path video data parallel transmission; Video component merging module: used to merge multiple received video component data according to the video block integrity information, and restore the video data that is closest to the original video frame.

Images