Block optimization method and device based on ultra-high-definition video transmission and medium

Abstract

The invention discloses a block optimization method and device based on ultra-high-definition video transmission and a medium, and relates to the technical field of ultra-high-definition video transmission, and the method comprises the steps: obtaining an ultra-high-definition video frame, and carrying out the video scene recognition of the ultra-high-definition video frame, so as to obtain a dynamic block suggestion image; based on the dynamic blocking suggestion graph, determining a video blocking result through blocking decision analysis; performing hierarchical quantization coding processing on the video partitioning result to obtain dynamic allocation data of differential quantization parameters; determining anti-fluctuation bandwidth data through bandwidth period prediction according to the dynamic allocation data of the differentiated quantization parameters; and based on the anti-fluctuation bandwidth data, code rate allocation data is obtained through dynamic code rate control. According to the method, the technical problems of large code rate fluctuation and obvious image quality loss of ultra-high-definition video block coding in the prior art are solved, the transmission stability under a fluctuation network is improved, and the performance bottleneck of the existing standard in 8K real-time transmission is broken through.

Description

Technical Field

[0001] The present application relates to the field of ultra-high-definition video transmission technology, and in particular to a block optimization method, device and medium based on ultra-high-definition video transmission. Background Art

[0002] With the advancement of video acquisition technology, ultra-high-definition video (such as 8K / 120fps) has become increasingly popular, resulting in a huge amount of video data. Traditional coding standards (such as HEVC) face the problems of high bandwidth occupancy and encoding and decoding delay in real-time transmission.

[0003] The existing mainstream block encoding scheme uses a fixed block size, which is difficult to adapt to complex dynamic scenes (such as fast-moving scenes), resulting in large bit rate fluctuations and obvious image quality loss. In addition, the block strategy problem in dynamic scenes has not been solved. When the network bandwidth fluctuates, the adaptive algorithm responds slowly, which is prone to freezes or resolution degradation. The existing technology cannot adjust the bit rate based on historical bandwidth predictions, and at the same time, it cannot combine video content features for targeted encoding optimization. Summary of the invention

[0004] The embodiments of the present application provide a block optimization method, device and medium based on ultra-high-definition video transmission, which solves the technical problems of large bit rate fluctuation and obvious image quality loss in ultra-high-definition video block encoding in the prior art.

[0005] In a first aspect, an embodiment of the present application provides a block optimization method based on ultra-high-definition video transmission, characterized in that the method includes: obtaining an ultra-high-definition video frame, and performing video scene recognition on the ultra-high-definition video frame to obtain a dynamic block suggestion map; based on the dynamic block suggestion map, determining the video block result through block decision analysis; performing layered quantization encoding processing on the video block result to obtain dynamic allocation data of differentiated quantization parameters; according to the dynamic allocation data of differentiated quantization parameters, determining anti-fluctuation bandwidth data through bandwidth period prediction; based on the anti-fluctuation bandwidth data, obtaining bit rate allocation data through dynamic bit rate control.

[0006] In one implementation of the present application, video scene recognition is performed on ultra-high-definition video frames to obtain a dynamic block recommendation map, specifically including: performing motion vector analysis on the ultra-high-definition video frames to obtain a first dynamic feature; performing texture complexity analysis on the ultra-high-definition video frames to obtain a second dynamic feature; determining a dynamic feature vector of the ultra-high-definition video frame based on the first dynamic feature and the second dynamic feature; and obtaining a dynamic block recommendation map through dynamic block analysis based on the dynamic feature vector; wherein the dynamic block analysis includes: region recommendation block size marking, block size optimization, and texture complexity scoring.

[0007] In one implementation of the present application, based on a dynamic block suggestion map, a video block result is determined through block decision analysis, specifically including: configuring block parameters of the dynamic block suggestion map to obtain a block video frame to be optimized; wherein the block parameter configuration includes: motion sensitive areas, complex texture areas, and continuous static areas; performing block boundary optimization on the block video frame to be optimized to determine the video block result; wherein the block boundary optimization includes: morphological closing operations and block size gradient change constraints.

[0008] In one implementation of the present application, after optimizing the block boundaries of the block video frames to be optimized to determine the video block results, the method also includes: determining the video adjustment mode based on the video block results through bandwidth change threshold analysis; wherein the video adjustment mode includes: normal mode and emergency mode; when the video adjustment mode is normal mode, maintaining the original video block state; when the video adjustment mode is emergency mode, dynamically adjusting the regional quality to obtain a dynamically optimized image; inserting a preset I frame into the dynamically optimized image to determine the emergency mode optimization data.

[0009] In one implementation of the present application, the video segmentation results are subjected to hierarchical quantization encoding processing to obtain dynamic allocation data of differential quantization parameters, specifically including: based on the video segmentation results, determining the motion level corresponding to the segmentation type through motion level matching; matching the motion level with the corresponding quantization parameter to obtain hierarchical quantization data; and determining the dynamic allocation data of differential quantization parameters through bit rate weight matching based on the hierarchical quantization data.

[0010] In one implementation of the present application, according to the dynamic allocation data of the differentiated quantization parameter, the anti-fluctuation bandwidth data is determined by bandwidth cycle prediction, specifically including: based on the dynamic allocation data, the current available bandwidth is obtained by periodic bandwidth detection; wherein the calculation formula of the periodic bandwidth detection is:

[0011] in, is the packet payload, It is The round trip time of a packet, is the current available bandwidth value; the current available bandwidth is subjected to anti-interference smoothing processing to obtain anti-fluctuation bandwidth data; wherein the calculation formula for the anti-interference smoothing processing is:

[0012] in, is the bandwidth value after bandwidth smoothing, The bandwidth value calculated last time.

[0013] In one implementation of the present application, based on the anti-fluctuation bandwidth data, the bit rate allocation data is obtained through dynamic bit rate control, which specifically includes: performing a bit rate dynamic calculation on the anti-fluctuation bandwidth data to obtain a bit rate to be allocated; wherein the calculation formula of the bit rate dynamic calculation is:

[0014] in, is the maximum bit rate supported by the encoder, For Block The code rate weight, For Block Current value; allocate the code rate to be allocated to the corresponding code to obtain code rate allocation data.

[0015] In one implementation of the present application, after obtaining the bit rate allocation data through dynamic bit rate control based on the anti-fluctuation bandwidth data, the method also includes: obtaining an ultra-high-definition optimized video based on the bit rate allocation data; transmitting the ultra-high-definition optimized video to a video destination through a preset transmission network to obtain an output ultra-high-definition video.

[0016] In the second aspect, an embodiment of the present application also provides a block optimization device based on ultra-high-definition video transmission, characterized in that the device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: obtain ultra-high-definition video frames, and perform video scene recognition on the ultra-high-definition video frames to obtain dynamic block suggestion maps; based on the dynamic block suggestion maps, determine the video block results through block decision analysis; perform hierarchical quantization encoding processing on the video block results to obtain dynamic allocation data of differentiated quantization parameters; based on the dynamic allocation data of differentiated quantization parameters, determine anti-fluctuation bandwidth data through bandwidth period prediction; based on the anti-fluctuation bandwidth data, obtain bit rate allocation data through dynamic bit rate control.

[0017] In the third aspect, the embodiment of the present application also provides a non-volatile computer storage medium based on block optimization of ultra-high-definition video transmission, storing computer executable instructions, characterized in that the computer executable instructions are set to: obtain ultra-high-definition video frames, and perform video scene recognition on the ultra-high-definition video frames to obtain a dynamic block suggestion map; based on the dynamic block suggestion map, determine the video block result through block decision analysis; perform layered quantization encoding processing on the video block result to obtain dynamic allocation data of differentiated quantization parameters; according to the dynamic allocation data of differentiated quantization parameters, determine the anti-fluctuation bandwidth data through bandwidth period prediction; based on the anti-fluctuation bandwidth data, obtain the bit rate allocation data through dynamic bit rate control.

[0018] The embodiments of the present application provide a block optimization method, device and medium based on ultra-high-definition video transmission. Through block decision and quantization, bit rate adjustment based on bandwidth prediction and real-time allocation of bit rate, the technical problems of large bit rate fluctuation and obvious image quality loss in ultra-high-definition video block encoding in the prior art are solved, dynamic encoding of ultra-high-definition video is realized, the stability of transmission under fluctuating networks is improved, the transmission delay is reduced, and the performance bottleneck of existing standards in 8K real-time transmission is broken through. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 A flow chart of a block optimization method based on ultra-high-definition video transmission provided in an embodiment of the present application; Figure 2 A schematic diagram of the internal structure of a block optimization device based on ultra-high-definition video transmission provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0021] The embodiments of the present application provide a block optimization method, device and medium based on ultra-high-definition video transmission. Through block decision and quantization, bit rate adjustment based on bandwidth prediction and real-time allocation of bit rate, the technical problems of large bit rate fluctuation and obvious image quality loss in ultra-high-definition video block encoding in the prior art are solved, dynamic encoding of ultra-high-definition video is realized, the stability of transmission under fluctuating networks is improved, the transmission delay is reduced, and the performance bottleneck of existing standards in 8K real-time transmission is broken through.

[0022] The technical solution proposed in the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0023] Figure 1 The following is a flow chart of a block optimization method based on ultra-high-definition video transmission provided in an embodiment of the present application. Figure 1 As shown, the embodiment of the present application provides a block optimization method based on ultra-high-definition video transmission, which specifically includes the following steps: Step 101: obtain an ultra-high-definition video frame, and perform video scene recognition on the ultra-high-definition video frame to obtain a dynamic block suggestion map.

[0024] For example, a pre-trained lightweight convolutional neural network (CNN) model is used to analyze the motion vector and texture complexity of the input video frame, and the dynamic feature vector of each frame is output, which is recorded as The lightweight convolutional neural network model contains 3 convolutional layers and 1 long short-term memory network (LSTM) layer. The model parameter volume is ≤1M, and the single frame processing time is ≤5ms, which realizes the dynamic feature analysis of ultra-high-definition video frames.

[0025] Specifically, video scene recognition is performed on ultra-high definition video frames to obtain a dynamic block suggestion map, including: performing motion vector analysis on the ultra-high definition video frames to obtain a first dynamic feature; performing texture complexity analysis on the ultra-high definition video frames to obtain a second dynamic feature; determining a dynamic feature vector of the ultra-high definition video frame based on the first dynamic feature and the second dynamic feature; and obtaining a dynamic block suggestion map through dynamic block analysis based on the dynamic feature vector; wherein the dynamic block analysis includes: region suggestion block size marking, block size optimization, and texture complexity scoring.

[0026] In one embodiment, a lightweight CNN model is first constructed, including the following four layers: Convolutional layer 1: 64 channels, 3×3 convolution kernel, stride 2, output resolution reduced to 1 / 4; Convolutional layer 2: 128 channels, 3×3 convolution kernel, stride 1, output feature map size unchanged; Convolutional layer 3: 256 channels, 1×1 convolution kernel, generating spatial feature vectors; LSTM layer: processes the features of 5 consecutive frames and predicts the changing trend of dynamic areas.

[0027] The database for model training is a preset 8K video library (including 10 types of scenes such as sports, medical imaging, etc.). Motion vector fields (MVF) are constructed for original video frames (YUV420 format, resolution ≥7680×4320).

[0028] The above model is used to process and output a dynamic block proposal map (marking the recommended block size of each area: 8×8 / 16×16 / 32×32 / 64×64 / 128×128) and a texture complexity score (0.0~1.0, calculated based on local variance).

[0029] Furthermore, the training data of the CNN model contains at least 100,000 frames of annotated 8K video samples, and the annotation information includes motion area masks and texture level labels. Low-precision quantization technology is used during model inference to convert floating-point weights into 8-bit integers, thereby reducing memory usage. The actual application reduction is about 60%.

[0030] Step 102: Based on the dynamic segmentation suggestion graph, determine the video segmentation result through segmentation decision analysis.

[0031] For example, since blocks in different states in ultra-high-definition video have different levels of attention from the user's perspective, in practical applications it is necessary to adjust the block parameters according to the characteristics of the blocks. The block features divide the video frames into multiple non-uniform coding blocks based on dynamic feature vectors, etc., and the block size is dynamically adjusted between 8×8 and 128×128.

[0032] Furthermore, the block rule is: the area with motion vector amplitude > threshold Th1 (for example: Th1=10 pixels / frame) adopts the minimum block size, and the area with texture complexity > threshold Th2 (for example, Th2=0.6) adopts the medium block size.

[0033] Specifically, based on the dynamic block suggestion map, the video block result is determined through block decision analysis, including: configuring the block parameters of the dynamic block suggestion map to obtain the block video frame to be optimized; wherein the block parameter configuration includes: motion sensitive area, complex texture area, continuous static area; performing block boundary optimization on the block video frame to be optimized to determine the video block result; wherein the block boundary optimization includes: morphological closing operation, block size gradient change constraint.

[0034] In one embodiment, different blocking strategies are used for different types of regions. Motion-sensitive regions (MVF amplitude > 10 pixels / frame) are forced to use the minimum blocking (8×8 or 16×16); complex texture regions (local variance > 0.7) use medium blocking (32×32 or 64×64); continuous static regions (MVF < 2 pixels / frame and variance < 0.3) are merged into 128×128 large blocks.

[0035] While dividing the blocks, the boundaries are optimized and the morphological closing operation is used to smooth the edges of the blocks to avoid jagged boundaries.

[0036] It should be noted that the method adopted is to limit the gradient of block size change, and the difference in adjacent block sizes does not exceed 2 levels (for example: direct transition from 32×32 to 128×128 is not allowed).

[0037] Furthermore, in the dynamic block decision, the Markov decision process (MDP) is used to optimize the block size, and the reward function is explained by the following formula: (1) in, =0.6, =0.4, both are weight coefficients, For image quality gain, The current available bandwidth value.

[0038] Furthermore, after optimizing the block boundaries of the block video frames to be optimized to determine the video block results, the method also includes: determining the video adjustment mode based on the video block results through bandwidth change threshold analysis; wherein the video adjustment modes include: normal mode and emergency mode; when the video adjustment mode is normal mode, maintaining the original video block state; when the video adjustment mode is emergency mode, dynamically adjusting the regional quality to obtain a dynamically optimized image; inserting a preset I frame into the dynamically optimized image to determine the emergency mode optimization data.

[0039] In one embodiment, when the video adjustment mode is the normal mode, the original video segmentation state is maintained, and when the bandwidth drops by more than 20%, the emergency mode is activated.

[0040] Increase the QP value of the static area by 10-15 while keeping the QP value of the moving area unchanged; and trigger the fast insertion of I frames, and the GOP length can be shortened to 60 frames.

[0041] Step 103: Perform hierarchical quantization encoding processing on the video segmentation result to obtain dynamic allocation data of differential quantization parameters.

[0042] For example, different quantization parameters (QP values) are allocated to different block types, with the QP value range of the motion area being 20-30 and the QP value range of the static area being 35-45. The H.266 / AVS3 standard is then used for encoding, and the bit rate allocation is controlled by the rate-distortion optimization (RDO) algorithm.

[0043] Specifically, the video segmentation results are subjected to hierarchical quantization encoding processing to obtain dynamic allocation data of differential quantization parameters, including: based on the video segmentation results, determining the motion level corresponding to the segmentation type through motion level matching; matching the motion level with the corresponding quantization parameter to obtain hierarchical quantization data; and according to the hierarchical quantization data, determining the dynamic allocation data of differential quantization parameters through bit rate weight matching.

[0044] In one embodiment, the motion level corresponding to the block type is determined by motion level matching; the motion level is matched with the corresponding quantization parameter to obtain hierarchical quantization data; according to the hierarchical quantization data, the dynamic allocation data of the differential quantization parameter is determined by bit rate weight matching to achieve dynamic adjustment of the QP value. The matching rules are explained in detail by the following table:

[0045] Table 1 Table 1 shows the QP value range corresponding to different blocks and motion levels in bit rate weight matching, so as to realize dynamic adjustment of QP value.

[0046] It should be noted that the QP value, or quantization parameter, is the abbreviation of Quantizer Parameter. In video encoding, it is a parameter that reflects the compression of spatial details. The smaller the QP value, the finer the quantization, the higher the image quality, the higher the corresponding bit rate, and the greater the transmission bandwidth consumption; conversely, the larger the QP value, the loss of some details during the encoding process, the image distortion, the quality degradation, the lower the bit rate, and the less transmission bandwidth consumption.

[0047] In this application, the QP value range adopted by the motion area is 20-30, and the QP value range adopted by the static area is 35-45. Then, the H.266 / AVS3 standard is adopted for encoding, and the allocation of the bit rate is controlled by the Rate-distortion optimization (RDO) algorithm.

[0048] Step 104: Determine the anti-fluctuation bandwidth data through bandwidth cycle prediction according to the dynamic allocation data of the differentiated quantization parameters.

[0049] Exemplarily, the network round-trip delay (RTT) and packet loss rate are collected every 50 ms to predict the available bandwidth. In order to reduce the impact of rapid fluctuations in bandwidth in a short period of time and avoid the discomfort caused by too frequent switching of encoding parameters, an exponential weighting method is used to smooth the bandwidth, thereby achieving ultra-high-definition video's resistance to rapid fluctuations in bandwidth in a short period of time and improving the stability of video transmission.

[0050] Specifically, according to the dynamic allocation data of the differentiated quantization parameter, the anti-fluctuation bandwidth data is determined through bandwidth cycle prediction, including: based on the dynamic allocation data, the current available bandwidth is obtained through cycle bandwidth detection; wherein the calculation formula of the cycle bandwidth detection is:

[0051] in, is the packet payload, It is The round trip time of a packet, is the current available bandwidth value; the current available bandwidth is subjected to anti-interference smoothing processing to obtain anti-fluctuation bandwidth data; wherein the calculation formula for the anti-interference smoothing processing is:

[0052] in, is the bandwidth value after bandwidth smoothing, The bandwidth value calculated last time.

[0053] In one embodiment, a group of detection packets are sent at fixed time intervals (eg, 50 ms), and the periodic bandwidth detection is explained by the following formula: (2) in, is the packet payload, It is The round-trip time (RTT) of a data packet. is the current available bandwidth value; the current available bandwidth is subjected to anti-interference smoothing processing to obtain anti-fluctuation bandwidth data, and max() is a maximum value function.

[0054] In order to reduce the impact of rapid bandwidth fluctuations in a short period of time and avoid the discomfort caused by too frequent switching of encoding parameters, bandwidth smoothing is performed using an exponential weighting method. The specific formula is as follows: The anti-interference smoothing process is explained by the following formula: (3) in, is the bandwidth value after bandwidth smoothing, is the bandwidth value calculated last time, The current available bandwidth value.

[0055] For weak network environments, packet loss compensation is also required. If the packet loss rate is greater than 5%, redundant coding is activated.

[0056] Step 105: Based on the anti-fluctuation bandwidth data, the bit rate allocation data is obtained through dynamic bit rate control.

[0057] For example, after determining the bandwidth value after bandwidth smoothing processing, in order to achieve the actual bit rate requirement of ultra-high-definition video under the bandwidth, it is necessary to dynamically control the bit rate.

[0058] Specifically, based on the anti-fluctuation bandwidth data, the bit rate allocation data is obtained through dynamic bit rate control, including: performing bit rate dynamic calculation on the anti-fluctuation bandwidth data to obtain the bit rate to be allocated; wherein the calculation formula of the bit rate dynamic calculation is:

[0059] in, is the maximum bit rate supported by the encoder, For Block The code rate weight, For Block Current value, is the code rate to be allocated, is the bandwidth value after bandwidth smoothing processing; the bit rate to be allocated is allocated to the corresponding encoding to obtain the bit rate allocation data.

[0060] Furthermore, after obtaining the bit rate allocation data through dynamic bit rate control based on the anti-fluctuation bandwidth data, the method also includes: obtaining ultra-high-definition optimized video based on the bit rate allocation data; transmitting the ultra-high-definition optimized video to a video destination through a preset transmission network to obtain output ultra-high-definition video.

[0061] In one embodiment, dynamic rate control is explained by the following formula: (4) in, is the maximum bit rate supported by the encoder, For Block The code rate weight, For Block Current value, is the code rate to be allocated. Furthermore, the encoding bitrate target value is dynamically adjusted according to the current available bandwidth value and video content characteristics. The target bitrate adjustment is explained by the following formula: (5) in, is the weight coefficient, which is generally taken as 0.7; is the bit rate to be allocated, i.e. the target bit rate, is the current available bandwidth value, is the historical average bit rate.

[0062] At the same time, high priority tags are assigned to the motion area coding blocks, which are transmitted first when the network is congested.

[0063] In one embodiment, through the above technical solution, an 8K original video (7680×4320@60fps, YUV420 format) is input, and the scene recognition module detects the presence of a fast-moving football and a static audience seat in the picture.

[0064] The coding blocks are dynamically divided into 16×16 blocks (QP=28) for the football area and 64×64 blocks (QP=35) for the audience area. When it is detected that the network bandwidth drops from 80Mbps to 30Mbps, the bitrate redistribution is initiated to prioritize the bitrate of the sports area and increase the QP value of the static area to 40.

[0065] The video output bit rate is stable at 25-28Mbps, the picture is complete and smooth, and the end-to-end delay is <150ms.

[0066] The above is an embodiment of the method proposed in this application. Based on the same inventive concept, the embodiment of this application also provides a block optimization device based on ultra-high-definition video transmission, whose structure is as follows: Figure 2 shown.

[0067] Figure 2 The following is a schematic diagram of the internal structure of a block optimization device based on ultra-high-definition video transmission provided in an embodiment of the present application. Figure 2 As shown, the device includes: at least one processor 201; and, a memory 202 communicatively connected to the at least one processor; The memory 202 stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor 201 to enable at least one processor 201 to: Acquire ultra-high-definition video frames and perform video scene recognition on the ultra-high-definition video frames to obtain a dynamic block suggestion map; based on the dynamic block suggestion map, determine the video block result through block decision analysis; perform hierarchical quantization encoding processing on the video block result to obtain dynamic allocation data of differentiated quantization parameters; determine anti-fluctuation bandwidth data through bandwidth cycle prediction based on the dynamic allocation data of differentiated quantization parameters; based on the anti-fluctuation bandwidth data, obtain bit rate allocation data through dynamic bit rate control.

[0068] Some embodiments of the present application provide corresponding Figure 1 A non-volatile computer storage medium based on block optimization of ultra-high-definition video transmission, storing computer executable instructions, wherein the computer executable instructions are set as: Acquire ultra-high-definition video frames and perform video scene recognition on the ultra-high-definition video frames to obtain a dynamic block suggestion map; based on the dynamic block suggestion map, determine the video block result through block decision analysis; perform hierarchical quantization encoding processing on the video block result to obtain dynamic allocation data of differentiated quantization parameters; determine anti-fluctuation bandwidth data through bandwidth cycle prediction based on the dynamic allocation data of differentiated quantization parameters; based on the anti-fluctuation bandwidth data, obtain bit rate allocation data through dynamic bit rate control.

[0069] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the IoT device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0070] The system and medium provided in the embodiments of the present application correspond one-to-one to the method. Therefore, the system and medium also have similar beneficial technical effects to the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the system and medium will not be repeated here.

[0071] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0072] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0073] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0074] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0075] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0076] Memory may include non-permanent storage in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0077] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0078] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0079] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A block optimization method based on ultra-high-definition video transmission, characterized in that: The method comprises: Acquire an ultra-high-definition video frame, and perform video scene recognition on the ultra-high-definition video frame to obtain a dynamic block suggestion map; Based on the dynamic block suggestion graph, determine the video block result through block decision analysis; Performing hierarchical quantization encoding processing on the video segmentation result to obtain dynamic allocation data of differential quantization parameters; Determining anti-fluctuation bandwidth data by bandwidth cycle prediction according to the dynamic allocation data of the differentiated quantization parameters; Based on the anti-fluctuation bandwidth data, bit rate allocation data is obtained through dynamic bit rate control.

2. The block optimization method based on ultra-high-definition video transmission according to claim 1 is characterized in that: Performing video scene recognition on the ultra-high-definition video frame to obtain a dynamic block suggestion map specifically includes: Performing motion vector analysis on the ultra-high definition video frame to obtain a first dynamic feature; Performing texture complexity analysis on the ultra-high definition video frame to obtain a second dynamic feature; Determining a dynamic feature vector of the ultra-high definition video frame based on the first dynamic feature and the second dynamic feature; According to the dynamic feature vector, the dynamic block suggestion map is obtained through dynamic block analysis; wherein the dynamic block analysis includes: region suggestion block size marking, block size optimization, and texture complexity scoring.

3. The block optimization method based on ultra-high-definition video transmission according to claim 1 is characterized in that: Based on the dynamic block suggestion graph, the video block result is determined through block decision analysis, specifically including: The dynamic block suggestion map is configured with block parameters to obtain a block video frame to be optimized; wherein the block parameter configuration includes: a motion sensitive area, a complex texture area, and a continuous static area; The block boundary optimization is performed on the block video frame to be optimized to determine the video block result; wherein the block boundary optimization includes: morphological closing operation and block size gradient change constraint.

4. The block optimization method based on ultra-high-definition video transmission according to claim 3 is characterized in that: After performing block boundary optimization on the block video frame to be optimized to determine the video block result, the method further includes: Based on the video segmentation result, the video adjustment mode is determined by analyzing the bandwidth change threshold; wherein the video adjustment mode includes: normal mode and emergency mode; When the video adjustment mode is the normal mode, maintaining the original video segmentation state; When the video adjustment mode is the emergency mode, dynamically adjusting the regional quality to obtain a dynamically optimized image; A preset I frame is inserted into the dynamic optimization image to determine the emergency mode optimization data.

5. The block optimization method based on ultra-high-definition video transmission according to claim 1 is characterized in that: The video segmentation result is subjected to hierarchical quantization coding processing to obtain dynamic allocation data of differential quantization parameters, specifically including: Based on the video segmentation result, determining the motion level corresponding to the segmentation type through motion level matching; Matching the motion level with a corresponding quantization parameter to obtain hierarchical quantization data; According to the hierarchical quantization data, the dynamic allocation data of the differential quantization parameter is determined through rate weight matching.

6. The block optimization method based on ultra-high-definition video transmission according to claim 1 is characterized in that: According to the dynamic allocation data of the differentiated quantization parameters, the anti-fluctuation bandwidth data is determined by bandwidth cycle prediction, specifically including: Based on the dynamic allocation data, the current available bandwidth is obtained through periodic bandwidth detection; wherein the calculation formula for the periodic bandwidth detection is: in, is the packet payload, It is The round trip time of a packet, is the current available bandwidth value; The current available bandwidth is subjected to an anti-interference smoothing process to obtain the anti-fluctuation bandwidth data; wherein the calculation formula for the anti-interference smoothing process is: in, is the bandwidth value after bandwidth smoothing, The bandwidth value calculated last time.

7. The block optimization method based on ultra-high definition video transmission according to claim 1 is characterized in that: Based on the anti-fluctuation bandwidth data, the bit rate allocation data is obtained through dynamic bit rate control, which specifically includes: The code rate dynamic operation is performed on the anti-fluctuation bandwidth data to obtain the code rate to be allocated; wherein the calculation formula of the code rate dynamic operation is: in, is the maximum bit rate supported by the encoder, For Block The code rate weight, For Block Current value, is the code rate to be allocated; The code rate to be allocated is allocated to the corresponding code to obtain the code rate allocation data.

8. The block optimization method based on ultra-high definition video transmission according to claim 1, characterized in that: After obtaining the bit rate allocation data through dynamic bit rate control based on the anti-fluctuation bandwidth data, the method further includes: Based on the bit rate allocation data, an ultra-high definition optimized video is obtained; The ultra-high-definition optimized video is transmitted to a video destination via a preset transmission network to obtain an output ultra-high-definition video.

9. A block optimization device based on ultra-high-definition video transmission, characterized in that: The device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: Acquire an ultra-high-definition video frame, and perform video scene recognition on the ultra-high-definition video frame to obtain a dynamic block suggestion map; Based on the dynamic block suggestion graph, determine the video block result through block decision analysis; Performing hierarchical quantization encoding processing on the video segmentation result to obtain dynamic allocation data of differential quantization parameters; Determining anti-fluctuation bandwidth data by bandwidth cycle prediction according to the dynamic allocation data of the differentiated quantization parameters; Based on the anti-fluctuation bandwidth data, bit rate allocation data is obtained through dynamic bit rate control.

10. A non-volatile computer storage medium based on block optimization for ultra-high-definition video transmission, storing computer executable instructions, characterized in that: The computer executable instructions are configured to: Acquire an ultra-high-definition video frame, and perform video scene recognition on the ultra-high-definition video frame to obtain a dynamic block suggestion map; Based on the dynamic block suggestion graph, determine the video block result through block decision analysis; Performing hierarchical quantization encoding processing on the video segmentation result to obtain dynamic allocation data of differential quantization parameters; Determining anti-fluctuation bandwidth data by bandwidth cycle prediction according to the dynamic allocation data of the differentiated quantization parameters; Based on the anti-fluctuation bandwidth data, bit rate allocation data is obtained through dynamic bit rate control.

Images