Video data processing method and device and video data processing system

Abstract

The invention provides a video data processing method and device and a video data processing system. The method comprises the following steps: acquiring video data of a coal mine; performing block processing on the video data to obtain a plurality of video blocks; compressing all the video blocks to obtain a plurality of compressed video blocks; combining all the compressed video blocks to obtain compressed video data; and sending the compressed video data to the target terminal. According to the scheme, the problem of low transmission efficiency of coal mine monitoring video data in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the technical field of coal mine safety monitoring, and in particular to a method and device for processing video data, a computer program product, and a video data processing system. Background Art

[0002] As the importance of coal mine safety production becomes increasingly prominent, coal mine monitoring systems have become an important means of coal mine safety production. By equipping each production link of coal mines with industrial monitoring systems, a complete set of information and automated network systems is formed, thereby greatly improving the safety management level of coal mine enterprises.

[0003] In order to ensure the stable operation of various systems on the security emergency platform, it is necessary to establish a high-speed and stable transmission link between the systems to ensure the fast and stable transmission of various types of information, especially video services.

[0004] Due to the complex environment inside the mine, the large amount of coal mine monitoring video data, high transmission and storage costs, and high real-time requirements, the current transmission efficiency of coal mine monitoring video data is relatively low. Summary of the invention

[0005] The main purpose of the present application is to provide a video data processing method, device, computer program product and video data processing system to at least solve the problem of low transmission efficiency of coal mine monitoring video data in the prior art.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for processing video data is provided, including: acquiring video data of a coal mine; performing block processing on the video data to obtain multiple video blocks; performing compression processing on all the video blocks to obtain multiple compressed video blocks; merging all the compressed video blocks to obtain compressed video data; and sending the compressed video data to a target terminal.

[0007] Optionally, obtaining video data of a coal mine includes: obtaining initial video data of the coal mine; and preprocessing the initial video data to obtain the video data, wherein the preprocessing method includes denoising and / or enhancement.

[0008] Optionally, the video data is processed into blocks to obtain a plurality of video blocks, including: obtaining fixed unit information, wherein the fixed unit information is information of a predefined data block size; and according to the fixed unit information, the video data is processed into blocks to obtain the plurality of video blocks.

[0009] Optionally, the video data is processed into blocks to obtain a plurality of video blocks, including: using image recognition technology to perform content analysis on the video data to obtain a recognition result, wherein the recognition result at least includes a region type and / or a richness of region details, and the region type includes a dynamic type and / or a static type; based on the recognition result, the video data is processed into blocks to obtain a plurality of the video blocks, wherein the video blocks divided by the dynamic type of video content are larger than the video blocks divided by the static type of video content, and the richness of the region details is positively correlated with the size of the video block.

[0010] Optionally, all the video blocks are compressed to obtain a plurality of compressed video blocks, including: calculating the Bhattacharyya coefficients of the Nth video block and the N+1th video block, wherein N≥1; determining the importance between the Nth video block and the N+1th video block according to the Bhattacharyya coefficients, wherein the magnitude of the Bhattacharyya coefficient is positively correlated with the importance; allocating sampling rates to all the video blocks according to the importance, and compressing the video blocks according to the sampling rates to obtain a plurality of compressed video blocks, wherein the importance is positively correlated with the sampling rate.

[0011] Optionally, all the video blocks are compressed to obtain multiple compressed video blocks, including: constructing a video compression model, wherein the video compression model is trained using multiple sets of training data, each set of training data in the multiple sets of training data includes historical video blocks obtained within a historical time period and historical compressed video blocks corresponding to the historical video blocks; inputting the video blocks into the video compression model to obtain the compressed video blocks corresponding to the video blocks.

[0012] Optionally, after sending the compressed video data to the target terminal, the method further includes: mirroring the image block of the compressed video data to obtain an extended reference frame; generating a search window in the extended reference frame, wherein the size of the search window is positively correlated with the inter-frame motion amplitude of the compressed video data; extracting the target image blocks that best match the current image block in the search window in turn by rasterization traversal to obtain reconstructed video data; decompressing the reconstructed video data to obtain multiple decompressed video blocks; and merging the multiple decompressed video blocks according to their positions during encoding to obtain decoded video data.

[0013] According to another aspect of the present application, a video data processing device is provided, including: an acquisition unit for acquiring video data of a coal mine; a blocking unit for performing block processing on the video data to obtain a plurality of video blocks; a compression unit for performing compression processing on all the video blocks to obtain a plurality of compressed video blocks; a first merging unit for merging all the compressed video blocks to obtain compressed video data; and a sending unit for sending the compressed video data to a target terminal.

[0014] According to another aspect of the present application, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the steps of any one of the video data processing methods.

[0015] According to another aspect of the present application, a video data processing system is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing any one of the video data processing methods.

[0016] By applying the technical solution of the present application, the video data of the coal mine can be decomposed into multiple small video blocks. In this way, when the video blocks are compressed, the compression speed is faster. Many small compressed video blocks after compression are merged into an overall compressed video data, making the video file smaller and the data sending speed faster, which can greatly improve the transmission efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 A hardware structure block diagram of a mobile terminal for executing a method for processing video data provided in an embodiment of the present application is shown;

[0019] Figure 2 A schematic flow chart of a method for processing video data provided according to an embodiment of the present application is shown;

[0020] Figure 3 A schematic flow chart of another method for processing video data is shown;

[0021] Figure 4 A structural block diagram of a video data processing device provided according to an embodiment of the present application is shown.

[0022] The above drawings include the following reference numerals:

[0023] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0024] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0025] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. 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 creative work should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] In an emergency environment, how to effectively compress coal mine monitoring video data, improve data transmission efficiency, reduce storage space, and ensure image quality has become an urgent problem to be solved. The use of traditional data encoding and decoding methods not only wastes a lot of sampling resources, but also makes the system overhead large under the condition of limited bandwidth resources in the underground communication environment, and it is also difficult to meet the requirements of fast and reliable transmission.

[0028] As introduced in the background technology, the transmission efficiency of coal mine monitoring video data in the prior art is low. To solve the above problems, the embodiments of the present application provide a video data processing method, device, computer program product and video data processing system.

[0029] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0030] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a method for processing video data according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0031] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the display method of device information in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] In this embodiment, a method for processing video data running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] Figure 2 FIG. 1 is a flow chart of a method for processing video data according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0034] Step S201, obtaining video data of a coal mine;

[0035] Specifically, multiple cameras can be installed underground to monitor the working environment and operation conditions of the coal mine, so as to obtain data of the video stream underground in the coal mine.

[0036] Step S202, dividing the video data into blocks to obtain a plurality of video blocks;

[0037] Specifically, continuous video frames can be divided into multiple smaller image blocks (video blocks). Since small blocks of data are easier to transmit, dividing video data into blocks can facilitate subsequent compression algorithms to operate independently on each block, and can also transmit video more quickly when bandwidth conditions are limited.

[0038] Step S203, compressing all the above video blocks to obtain multiple compressed video blocks;

[0039] Specifically, after the video block segmentation is completed, each video block is compressed. The purpose of compression is to reduce the size of video data so that it can be transmitted more efficiently under limited network bandwidth. A variety of algorithms can be used to compress video, such as block-based compressed sensing, Huffman coding, JPEG, H.264 / H.265 and other video coding standards, which can minimize the amount of data while maintaining video quality.

[0040] Step S204, merging all the above compressed video blocks to obtain compressed video data;

[0041] Specifically, all compressed video blocks are reassembled into a continuous video stream. The combined data will be compressed video data, which is significantly smaller than the original video data obtained in step S201, but can still maintain the key information and visual quality of the video.

[0042] Step S205: sending the compressed video data to the target terminal.

[0043] Specifically, the compressed video data is transmitted through the network to the designated receiving terminal, such as the server of the monitoring center, the mobile device used by the manager, etc. Through the real-time transmission protocol (such as RTSP, RTMP, WebRTC, etc.), the fast and stable transmission of video data is ensured to achieve remote real-time monitoring.

[0044] Through this embodiment, the video data of the coal mine can be decomposed into multiple small video blocks, so that when the video blocks are compressed, the compression speed is faster, and many small compressed video blocks are merged into an overall compressed video data, so that the video file is smaller and the data is sent faster, which can greatly improve the transmission efficiency.

[0045] Specifically, this solution can solve the drawbacks of existing encoding and decoding and transmission technologies. It must not only ensure the quality of monitoring videos so that managers can clearly see the scenes underground, but also ensure that the video size meets the requirements of fast and stable underground transmission as much as possible, so as to improve data transmission efficiency, reduce storage space and real-time performance, so that managers can monitor the dynamics underground in real time, thereby ensuring mine production safety and avoiding accidents as much as possible.

[0046] In order to improve the quality of data and obtain video data of a coal mine, the following steps can be taken: obtaining the initial video data of the coal mine; preprocessing the initial video data to obtain the video data, wherein the preprocessing method includes denoising and / or enhancement.

[0047] In this solution, through preprocessing, the original initial video data that may have quality problems can be converted into video data that is more suitable for compression and transmission, thereby improving the data quality.

[0048] Specifically, the initial video data of coal mine monitoring can be preprocessed, including denoising, enhancement and other operations, to improve the effects of subsequent compression and reconstruction.

[0049] Denoising: Coal mine monitoring videos often have noise interference due to environmental noise, equipment limitations, etc., and the video needs to be denoised. First, determine a filter window and position (usually containing an odd number of pixels), that is, determine a neighborhood with a certain pixel as the center point, which is generally a square neighborhood, but can also be a circle, a cross, etc.; then sort the grayscale values ​​of each pixel in the neighborhood, and finally take the median to replace the pixel value in the center of the original window. This operation can effectively remove noise from the video and improve the video quality. The denoising operation can use a variety of algorithms, such as median filtering, mean filtering, Gaussian filtering, bilateral filtering, etc., the purpose is to remove these noise interferences, make the video clearer, and improve the effect of subsequent compression and reconstruction.

[0050] Enhancement: The visual effects of coal mine monitoring videos may be affected by factors such as light and angle, and need to be enhanced. The enhancement method used in this application is: to perform non-homogeneous stretching on the original image to expand the pixel value spacing and uniformize the pixel quantity in each grayscale range. Enhancement processing can improve the visual effect of the video and improve the effect of subsequent compression and reconstruction. The visual effects of coal mine monitoring videos may be affected by problems such as insufficient light, low contrast, and color distortion. The enhancement operation can improve the visual quality of the video by adjusting image parameters such as brightness, contrast, and color saturation, or using more complex image processing techniques such as histogram equalization, sharpening, and chroma adjustment, making it easier to monitor and analyze.

[0051] Specifically, in the coal mine monitoring system, multiple high-definition cameras are installed in key areas underground, such as working faces, transportation tunnels, ventilation systems, etc., to capture video data in a 24-hour uninterrupted manner. The video data captured by these cameras has not been processed in any way and contains all the original information. We define this data as initial video data.

[0052] Specifically, an adaptive median filter algorithm is used to denoise the video. The algorithm can automatically adjust the filter window size according to the local environment of the pixel in the video, so as to remove noise more effectively while avoiding affecting the image details. For each video frame, the image is divided into multiple fixed-size regions (such as 8x8 pixels). In each region, the algorithm checks the grayscale value difference between the current pixel and its neighboring pixels. If the difference exceeds the preset threshold (indicating that it may be noise), a median filter operation is performed around the pixel; if the difference is small, the pixel value is kept unchanged. By adjusting the threshold and the filter window size, the denoising effect can be flexibly controlled to meet the video quality requirements of different scenes.

[0053] Specifically, histogram equalization and contrast enhancement techniques are used to improve the visual effect of the video, especially in underground environments with large light changes or insufficient light. By analyzing the grayscale histogram of the video frame, the distribution of pixel grayscale values ​​is adjusted to make the details of brighter or darker areas of the image more obvious. Using an adaptive contrast enhancement algorithm, the brightness and contrast of the image are adjusted according to the local contrast of the video frame to highlight the details of key parts. According to the lighting conditions of the underground environment, the parameters of histogram equalization and contrast enhancement are dynamically adjusted to achieve the best enhancement effect.

[0054] Specifically, the preprocessed coal mine monitoring video signal can be processed in blocks, and the coal mine monitoring video can be divided into multiple small blocks. The size of each block can be adjusted as needed. The following is divided into two cases for introduction.

[0055] During the specific implementation process, the above-mentioned video data is processed into blocks to obtain multiple video blocks, which can be achieved through the following steps: obtaining fixed unit information, wherein the above-mentioned fixed unit information is information of a predefined data block size; according to the above-mentioned fixed unit information, the above-mentioned video data is processed into blocks to obtain multiple video blocks.

[0056] In this solution, video data can be decomposed into more manageable units to facilitate subsequent video compression, encoding and transmission, which can further improve the efficiency of video compression and the real-time performance of transmission.

[0057] Specifically, the above solution is to divide the video frame into blocks of fixed size, such as 8*8 or 16*16 image blocks. This block division method is simple and easy to use.

[0058] Specifically, fixed unit information refers to parameters that have been set before video processing to guide how video data is divided into smaller blocks. In video coding, the choice of data block size directly affects the coding efficiency and the quality of the decoded image. Usually, the choice of data block size needs to be determined based on the resolution, frame rate, and expected compression ratio and transmission bandwidth of the video. For example, 8x8 pixels or 16x16 pixels are the most common block sizes, but other sizes can also be selected according to requirements. Obtaining fixed unit information is usually pre-set by the system or user before video processing begins.

[0059] Specifically, based on the obtained fixed unit information, the video data is divided into multiple small blocks, and the size of each block is consistent with the data block size defined in the fixed unit information. The purpose of block processing is to decompose the video data into more manageable units to facilitate subsequent video compression, encoding and transmission. Through block processing, compression algorithms can be applied independently to each video block, reducing the complexity of the overall video processing and improving transmission efficiency, especially when the network bandwidth is limited. The segmentation of video blocks can be performed based on continuous frames of video data or within each frame.

[0060] For example, assuming that the data block size defined by the fixed unit information is 16x16 pixels, when segmenting each frame of video data, the system's processing flow can be: Calculate the width and height of the video frame. Determine the starting and ending positions of each data block to ensure that the size of each data block is 16x16 pixels. Starting from the first row and first column of the video frame, segment out 16x16 pixel video blocks in sequence. If the width or height of the video frame is not divisible by 16, the data block of the last row or column may not be a complete 16x16 pixels, and special processing such as padding or cropping is required to keep the block size uniform. Repeat the above process until each part of the video frame is segmented into 16x16 pixel blocks.

[0061] In order to improve the efficiency and quality of video compression, the following optimization solutions can also be adopted:

[0062] 1. Dynamically adjust the data block size. Although the fixed unit information defines the initial size of the data block, during the video encoding process, the video content can be dynamically analyzed to use smaller data blocks for moving areas and larger data blocks for static areas to achieve better compression effect and image quality.

[0063] 2. Utilize the similarity of adjacent frames. In the block process, considering that adjacent frames usually have a high degree of similarity, the block information of the previous frame can be used to predict the block mode of the current frame, thereby reducing the computational complexity of the block operation.

[0064] 3. Parallel processing. For high-resolution videos, block processing can be performed in parallel, that is, multiple data blocks in the video frame are processed simultaneously, thereby speeding up the processing speed and being more suitable for real-time monitoring needs.

[0065] In a specific implementation process, the video data is divided into blocks to obtain a plurality of video blocks, which can be achieved through the following steps: using image recognition technology to perform content analysis on the video data to obtain a recognition result, wherein the recognition result at least includes the region type and / or the richness of the region details, and the region type includes a dynamic type and / or a static type; based on the recognition result, the video data is divided into blocks to obtain a plurality of the video blocks, wherein the video blocks divided by the video content of the dynamic type are larger than the video blocks divided by the video content of the static type, and the richness of the region details is positively correlated with the size of the video block.

[0066] In this solution, the division of video blocks can be intelligently adjusted according to the characteristics of the video content, thereby improving the compression efficiency while ensuring the integrity of key information and a high level of video quality.

[0067] Specifically, the above scheme is to divide the video frame into blocks of variable size, for example, dynamically adjust the block size according to the movement and position of the object in the frame. This block division method can better adapt to the changes in video content, but the amount of calculation will be greater than the first method.

[0068] Specifically, image recognition technology can analyze the content in the video frame and identify different types of regions and the richness of regional details. For example, by analyzing pixel changes, texture features, color distribution, etc. within the frame, dynamic areas (such as human activities, machine operation, etc.) and static areas (such as background, fixed equipment, etc.) can be distinguished.

[0069] Specifically, the recognition result includes at least the region type and / or the richness of the region details, which provides a basis for subsequent video segmentation. Dynamic regions may contain richer details due to activity, while static regions usually have less changes and lower detail richness.

[0070] Specifically, the block processing is performed based on the recognition result, that is, the size and number of the video blocks are determined according to the dynamic or static nature of the video content and the richness of details.

[0071] Specifically, the number of video blocks divided into dynamic video content is greater than that of static video content. This is because the information in the dynamic area is more complex, and finer block division can better capture dynamic changes and ensure accurate encoding of key information.

[0072] Specifically, the richness of regional details is positively correlated with the size of the video block, which means that the more detailed the area, the smaller the divided video block is for more detailed encoding, while the area with less details can be divided into larger blocks to reduce the amount of encoding calculation and data.

[0073] In addition, deep learning models, such as convolutional neural networks (CNN) or recurrent neural networks (RNN), can be used to deeply analyze video data to identify dynamic and static areas, as well as the richness of details in different areas. Feature extraction algorithms, such as SIFT (Scale Invariant Feature Transform), SURF (Speeded Up Robust Features) or HOG (Histogram Oriented Gradients), can be used to quantify the richness of regional details.

[0074] In addition, dynamic areas can be fine-blocked. For dynamic areas, due to their fast content changes and high detail richness, the system divides these areas into smaller video blocks, such as 8x8 pixels or smaller, in order to capture and encode dynamic information more finely. Static areas are block-blocked. On the contrary, static areas have less content changes and lower detail richness, and can be divided into larger video blocks, such as 16x16 pixels or larger, to reduce redundant information during the encoding process.

[0075] In addition, a threshold for detail richness can be set, and the size of the video block can be dynamically adjusted according to the threshold. For example, if the detail richness of a certain area exceeds a preset threshold, it is divided into smaller blocks, and vice versa, it can be divided into larger blocks. The numerical range of quantized richness can be 0 to 1, where 0 represents the least rich (such as a pure color area) and 1 represents the richest (such as an area with complex textures, multiple colors, high-density edges, and frequent motion). It can also be a more specific range, such as 1 to 100, where the higher the value, the higher the richness. The actual quantization method and numerical range depend on the analysis algorithm used and the specific application scenario.

[0076] In addition, during the video encoding process, the system can dynamically adjust the size of the video block by monitoring the content changes and detail richness of the video frame in real time to achieve optimal encoding efficiency and video quality.

[0077] In addition, for continuous video frames, the recognition result of the previous frame can be used to predict the blocking mode of the current frame, reducing unnecessary analysis and blocking operations and speeding up processing.

[0078] In addition, on a multi-core processor or GPU, different areas in a video frame, especially dynamic areas and areas with rich details, can be processed in parallel to improve the overall encoding efficiency.

[0079] In addition, during the encoding process, the system can dynamically adjust the size and encoding quality of the video block according to the real-time bandwidth of the network to ensure the transmission of the highest possible quality video data under limited bandwidth.

[0080] In some embodiments, all of the above-mentioned video blocks are compressed to obtain multiple compressed video blocks, which can be specifically achieved through the following steps: calculating the Bhattacharyya coefficients of the Nth video block and the N+1th video block, where N≥1; determining the importance between the Nth video block and the N+1th video block based on the Bhattacharyya coefficients, where the size of the Bhattacharyya coefficient is positively correlated with the importance; allocating sampling rates to all of the above-mentioned video blocks based on the importance, and compressing the above-mentioned video blocks based on the sampling rates to obtain multiple compressed video blocks, where the importance is positively correlated with the sampling rate.

[0081] In this solution, the changing characteristics of video content can be automatically identified, and the sampling rate and encoding strategy can be dynamically adjusted according to the degree of change, thereby effectively reducing the amount of data transmission while ensuring video quality, thereby further improving transmission efficiency.

[0082] Specifically, for each video block, its normalized Bhattacharyya coefficient can be calculated; according to the size of the normalized Bhattacharyya coefficient, the importance of each pixel in the video block can be determined; based on the importance of the video block, the sampling rate is allocated, and the obtained adaptive sampling number is combined with the block-based compressed sensing method to perform an adaptive compression process of the video image; the compressed video blocks are merged to obtain the final compressed video data.

[0083] Specifically, after the video frame is divided into small image blocks, the normalized Bhattacharyya coefficient (NBC) is calculated for each image block. The Bhattacharyya coefficient is used to approximate the overlap between two statistical samples, so it can be used to measure the correlation of samples. In coal mine monitoring videos, the statistical samples are usually the pixel values ​​of the image, and the size of the Bhattacharyya coefficient represents the similarity between images.

[0084] For each image block of a video frame, the NBC is calculated as follows:

[0085] Initialize a histogram array of size 256 for the current frame and the reference frame respectively, and then perform frequency distribution statistics based on pixel values. The histogram array of the reference frame is stored during the sampling process of the frame, and is discarded after the NBC calculation with the current frame. The histogram array of the current frame is used in the calculation of subsequent video frames. The histogram distributions of image blocks with high similarity have high similarity. Use the obtained histogram array to calculate the normalized Bhattacharyya coefficient.

[0086] Specifically, the importance of the video block is determined according to the size of the normalized Bhattacharyya coefficient. The NBC between different video blocks in the video sequence, the closer the NBC is to 0, the smaller the difference between the video blocks, and the lower the importance of the current video block relative to the reference video block.

[0087] Specifically, based on the importance of the video block, the sampling rate is allocated, and the obtained adaptive sampling number is combined with the block-based compressed sensing method to perform the adaptive compression process of the video image. The calculation of the NBC value is transferred from the frame to the block, and the adaptive sampling rate allocation step is performed according to the ratio between the NBCs of different image blocks, so as to allocate more sampling numbers to the image blocks with larger inter-frame motion. In particular, for the background part that does not change between frames, its NBC value will be 0. This feature is particularly suitable for application in the sampling compression of surveillance video images.

[0088] Specifically, the compressed video blocks are merged to obtain final compressed video data, which is transmitted via a real-time transport protocol.

[0089] Specifically, the Bhattacharyya Coefficient is a statistic used to measure the similarity between two probability distributions, with a value between 0 and 1. The closer the value is to 1, the more similar the two distributions are. In video processing, the calculation of the Bhattacharyya coefficient of consecutive video blocks (such as the Nth and N+1th blocks) is based on their pixel value distribution. Usually, this involves comparing the grayscale histogram or color histogram of the video blocks to quantify the similarity between them. The purpose of calculating the Bhattacharyya coefficient of video blocks N and N+1 is to evaluate the content changes between the two consecutive video blocks, so as to determine their relative importance in the video sequence.

[0090] Specifically, the importance between video blocks can be determined based on the calculated Bhattacharyya coefficient. Since the size of the Bhattacharyya coefficient is positively correlated with the importance, this means that if the Bhattacharyya coefficient is high (the similarity between two blocks is high), the relative importance between them is low, and if the Bhattacharyya coefficient is low (the similarity is low, that is, the difference is large), the relative importance between them is high. Importance reflects the amount of information contained in the video block and the degree of change, and is a key factor in determining how to sample and compress.

[0091] Specifically, different sampling rates are assigned to video blocks according to their importance. Video blocks with high importance (i.e., small Bhattacharyya coefficient and large variation) will be assigned higher sampling rates to ensure complete retention of details and high-quality reconstruction; video blocks with low importance (i.e., large Bhattacharyya coefficient and small variation) will be assigned lower sampling rates to reduce the amount of data and the size of the compressed file. The sampling rate is positively correlated with the importance, that is, the higher the importance, the higher the sampling rate, and vice versa. Each video block is compressed according to the assigned sampling rate to generate multiple compressed video blocks. The compression process is to minimize the amount of data transmission while ensuring the integrity of the video information.

[0092] In addition, the threshold of the Bhattacharyya coefficient can be adjusted according to the video content and environmental conditions to meet the compression requirements of different scenes. For example, in an environment with large light changes, the threshold can be set smaller to capture detail changes more sensitively. It is also possible to establish an optimization model between sampling rate, compression parameters and video quality through experiments or machine learning methods to achieve the best compression effect and transmission efficiency while ensuring video quality. When calculating the importance, not only the Bhattacharyya coefficient of two consecutive frames (N and N+1) is considered, but also a multi-frame prediction mechanism can be introduced to consider the similarity and importance of multiple consecutive frames such as N-1, N, N+1, etc., to more comprehensively evaluate the change trend of the video block.

[0093] In some embodiments, all of the above-mentioned video blocks are compressed to obtain multiple compressed video blocks, which can be specifically achieved through the following steps: constructing a video compression model, wherein the above-mentioned video compression model is trained using multiple sets of training data, and each set of training data in the above-mentioned multiple sets of training data includes historical video blocks obtained within a historical time period and historical compressed video blocks corresponding to the above-mentioned historical video blocks; inputting the above-mentioned video blocks into the above-mentioned video compression model to obtain the above-mentioned compressed video blocks corresponding to the above-mentioned video blocks.

[0094] In this solution, the deep learning-based video compression model can not only automate and optimize the video compression process, but also dynamically adjust the compression strategy according to the characteristics of the video content, such as motion changes, detail richness, etc., to ensure better compression efficiency.

[0095] Specifically, inputting a video block into a video compression model means using a trained model to compress the new video block. The model will analyze and optimize the input video block based on the learned compression strategy to generate the corresponding compressed video block. This process can be regarded as a prediction, that is, the model predicts the compressed version of the input video block, and this prediction is based on the rules learned by the model from multiple sets of training data during the training phase.

[0096] Specifically, a large amount of video data can be collected, including videos of different scenes, different resolutions, and different contents. These video data are divided into video blocks, and the original information of each block is recorded. At the same time, these blocks are compressed using traditional video compression algorithms (such as H.264, H.265, etc.) to generate compressed video blocks as the target output of training data. The preprocessing stage may also include denoising, standardization, and other operations on the video blocks to improve the training effect.

[0097] Specifically, a deep learning model can be designed, usually based on structures such as convolutional neural networks (CNN) or generative adversarial networks (GAN). The input of the model is the original video block, and the output is the compressed video block. This model needs to be trained using multiple sets of collected training data through methods such as backpropagation and gradient descent to minimize the difference between the predicted compressed block and the actual compressed block. The training process may involve adjusting the model's architecture, optimization algorithms, and hyperparameters to obtain the best compression performance.

[0098] Specifically, after training, an independent test dataset can be used to verify the performance of the model. The test dataset should not be included in the training dataset to ensure the objectivity of the results. The verification process may include evaluating indicators such as the visual quality, file size, compression ratio, and decompression speed of the compressed video blocks to ensure that the model has found a good balance between compression efficiency and decompression quality.

[0099] Specifically, once the model is trained and verified, it can be deployed in a real-time video compression system. The real-time system can receive the video stream, split it into blocks, and compress each block using the video compression model. In practical applications, it may be necessary to further optimize the operating efficiency of the model, such as through model pruning, quantization, or acceleration using dedicated hardware to ensure that it can meet real-time requirements.

[0100] In some embodiments, after the compressed video data is sent to the target terminal, the method further includes the following steps: mirroring the image block of the compressed video data to obtain an extended reference frame; generating a search window in the extended reference frame, wherein the size of the search window is positively correlated with the inter-frame motion amplitude of the compressed video data; extracting the target image blocks that best match the current image block in the search window in turn by rasterization traversal to obtain reconstructed video data; decompressing the reconstructed video data to obtain multiple decompressed video blocks; and merging the multiple decompressed video blocks according to their positions during encoding to obtain decoded video data.

[0101] In this solution, the reconstruction accuracy of video data is improved by mirror projection and search window, and the size of the search window is dynamically adjusted according to the inter-frame motion amplitude to better match the target image block. By decompressing and merging the decompressed video blocks, high-quality decoded video data is finally obtained. It can improve the compression rate and decompression quality of video data and is suitable for various scenarios of video encoding and decoding.

[0102] Specifically, the compressed video data is transmitted through a real-time transport protocol; in an emergency environment, appropriate video blocks are selected for reconstruction according to the video quality and real-time requirements that need to be reconstructed; the reconstructed coal mine monitoring video is post-processed, including denoising, enhancement and other operations, to improve image quality.

[0103] Specifically, in an emergency environment, appropriate video blocks are selected for reconstruction according to the video quality and real-time requirements that need to be reconstructed.

[0104] Step 1: Expand the reference frame. Image blocks at the edge of the image cannot generate a complete search window during reconstruction, so the size of the reference frame needs to be expanded. One feasible method is to mirror-project all image blocks at the edge to obtain an expanded reference frame.

[0105] Step 2: Generate a search window at the same position as the current image block in the expanded reference frame. The size of the search window is usually related to the inter-frame motion amplitude of the video. Video sequences with large motion amplitudes usually require a larger search window to improve the accuracy of the generated hypothesis set.

[0106] Step 3: In the search window, we obtain all image blocks by rasterization traversal, vectorize them to generate a search matrix, and then search for the best matching image block.

[0107] Step 4: Generate a hypothesis set generation window with the position of the searched matching block as the center, and perform rasterization traversal on the hypothesis set generation window to obtain a hypothesis set.

[0108] Step 5: Use the same method to obtain hypothesis sets in other reference frames.

[0109] The reconstructed coal mine monitoring video is post-processed, including denoising, enhancement and other operations, to improve the image quality.

[0110] Compared with general application scenarios, the lighting conditions in coal mines are poor and the environment is relatively harsh, so the resulting surveillance video images are closer to grayscale images. The transmission bandwidth between line video sensor nodes is relatively low, so the compression reconstruction algorithm needs to reconstruct high-quality video sequences at a low sampling rate while maintaining a high reconstruction efficiency, which places high demands on the algorithm's compression and reconstruction performance. This application solves the problems of the prior art very well.

[0111] In summary, the present scheme can effectively discover and utilize the spatiotemporal redundancy existing in the video sequence, so as to allocate a higher number of samples to the foreground area where inter-frame motion occurs, and allocate a lower number of samples to the background area, thereby effectively reducing unnecessary sampling of the background area of ​​the video image, and achieving good reconstruction effects in video images with a small inter-frame motion amplitude. This method is particularly suitable for video monitoring application scenarios, especially in wireless video sensor networks underground in coal mines. By adopting a method of dividing the video signal into blocks and combining it with a deep learning model, the coal mine monitoring video is compressed, the amount of data is reduced, and the storage space is reduced. At the same time, appropriate video blocks can be selected for reconstruction according to the needs of the emergency environment, thereby improving real-time performance. In addition, the present invention also adopts a reconstruction operation based on interpolation to improve image quality. Therefore, the present invention has good practicality and broad application prospects.

[0112] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the video data processing method of the present application will be described in detail below in conjunction with specific embodiments.

[0113] This embodiment relates to a specific method for processing video data, such as Figure 3 As shown, the following steps are included:

[0114] Step S1, preprocessing the coal mine monitoring video, including denoising, enhancement and other operations, to improve the effect of subsequent compression and reconstruction;

[0115] Step S2, performing block processing on the pre-processed coal mine monitoring video signal, dividing the coal mine monitoring video into a plurality of small blocks, and the size of each block can be adjusted as needed;

[0116] Step S3, for each video block, calculating its normalized Bhattacharyya coefficient;

[0117] Step S4, determining the importance of each pixel in the video block according to the size of the normalized Bhattacharyya coefficient;

[0118] Step S5, allocating sampling rates based on the importance of the video blocks, combining the obtained adaptive sampling number with the block-based compressed sensing method to perform an adaptive compression process of the video image;

[0119] Step S6, merging the compressed video blocks to obtain final compressed video data; transmitting the compressed video data via a real-time transport protocol;

[0120] Step S7, in an emergency environment, selecting appropriate video blocks for reconstruction according to the video quality and real-time requirements to be reconstructed;

[0121] Step S8, post-processing the reconstructed coal mine monitoring video, including denoising, enhancement and other operations, to improve image quality.

[0122] The embodiment of the present application also provides a kind of processing device of video data. It should be noted that the processing device of video data of the embodiment of the present application can be used to execute the processing method for video data provided by the embodiment of the present application. The device is used to implement the above-mentioned embodiment and preferred implementation mode, and the description has been made and will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.

[0123] The following is an introduction to the video data processing device provided in the embodiments of the present application.

[0124] Figure 4 is a structural block diagram of a video data processing device according to an embodiment of the present application. Figure 4 As shown, the device comprises:

[0125] An acquisition unit 10, used for acquiring video data of a coal mine;

[0126] A block division unit 20, configured to perform block processing on the video data to obtain a plurality of video blocks;

[0127] A compression unit 30, configured to compress all of the above video blocks to obtain a plurality of compressed video blocks;

[0128] A first merging unit 40 is used to merge all the above compressed video blocks to obtain compressed video data;

[0129] The sending unit 50 is used to send the compressed video data to the target terminal.

[0130] Through this embodiment, the video data of the coal mine can be decomposed into multiple small video blocks, so that when the video blocks are compressed, the compression speed is faster, and many small compressed video blocks are merged into an overall compressed video data, so that the video file is smaller and the data is sent faster, which can greatly improve the transmission efficiency.

[0131] In order to improve the quality of data, the acquisition unit includes a first acquisition module and a preprocessing module. The first acquisition module is used to acquire the initial video data of the above-mentioned coal mine; the preprocessing module is used to preprocess the above-mentioned initial video data to obtain the above-mentioned video data, wherein the preprocessing method includes denoising and / or enhancement.

[0132] In this solution, through preprocessing, the original initial video data that may have quality problems can be converted into video data that is more suitable for compression and transmission, thereby improving the data quality.

[0133] In the specific implementation process, the blocking unit includes a second acquisition module and a first blocking module. The second acquisition module is used to obtain fixed unit information, wherein the fixed unit information is information of a predefined data block size; the first blocking module is used to block the video data according to the fixed unit information to obtain multiple video blocks.

[0134] In this solution, video data can be decomposed into more manageable units to facilitate subsequent video compression, encoding and transmission, which can further improve the efficiency of video compression and the real-time performance of transmission.

[0135] During the specific implementation process, the blocking unit includes an analysis module and a second blocking module. The analysis module is used to use image recognition technology to perform content analysis on the above-mentioned video data to obtain a recognition result, wherein the above-mentioned recognition result at least includes the region type and / or the richness of the region details, and the above-mentioned region type includes a dynamic type and / or a static type; the second blocking module is used to block the above-mentioned video data according to the above-mentioned recognition result to obtain multiple above-mentioned video blocks, wherein the above-mentioned video blocks divided by the above-mentioned dynamic type of video content are larger than the above-mentioned video blocks divided by the above-mentioned static type of video content, and the richness of the region details is positively correlated with the size of the above-mentioned video block.

[0136] In this solution, the division of video blocks can be intelligently adjusted according to the characteristics of the video content, thereby improving the compression efficiency while ensuring the integrity of key information and a high level of video quality.

[0137] In some embodiments, the compression unit includes a calculation module, a determination module and a compression module, the calculation module is used to calculate the Bhattacharyya coefficient of the Nth video block and the N+1th video block, wherein N≥1; the determination module is used to determine the importance between the Nth video block and the N+1th video block according to the Bhattacharyya coefficient, wherein the size of the Bhattacharyya coefficient is positively correlated with the importance; the compression module is used to allocate sampling rates to all the video blocks according to the importance, and compress the video blocks according to the sampling rates to obtain multiple compressed video blocks, wherein the importance is positively correlated with the sampling rate.

[0138] In this solution, the changing characteristics of video content can be automatically identified, and the sampling rate and encoding strategy can be dynamically adjusted according to the degree of change, thereby effectively reducing the amount of data transmission while ensuring video quality, thereby further improving transmission efficiency.

[0139] In some embodiments, the compression unit includes a construction module and a processing submodule, the construction module is used to construct a video compression model, wherein the above-mentioned video compression model is trained using multiple sets of training data, each set of training data in the above-mentioned multiple sets of training data includes historical video blocks obtained within a historical time period and historical compressed video blocks corresponding to the above-mentioned historical video blocks; the processing submodule is used to input the above-mentioned video blocks into the above-mentioned video compression model to obtain the above-mentioned compressed video blocks corresponding to the above-mentioned video blocks.

[0140] In this solution, the deep learning-based video compression model can not only automate and optimize the video compression process, but also dynamically adjust the compression strategy according to the characteristics of the video content, such as motion changes, detail richness, etc., to ensure better compression efficiency.

[0141] In some embodiments, the above-mentioned device also includes a projection unit, a generation unit, a traversal unit, a decompression unit and a second merging unit, the projection unit is used to mirror the image block of the above-mentioned compressed video data after sending the above-mentioned compressed video data to the target terminal to obtain an extended reference frame; the generation unit is used to generate a search window in the above-mentioned extended reference frame, wherein the size of the above-mentioned search window is positively correlated with the inter-frame motion amplitude of the above-mentioned compressed video data; the traversal unit is used to extract the target image blocks that best match the current image block in the above-mentioned search window in turn by rasterization traversal to obtain reconstructed video data; the decompression unit is used to decompress the above-mentioned reconstructed video data to obtain multiple decompressed video blocks; the multiple decompressed video blocks are merged according to the positions at the time of encoding to obtain decoded video data.

[0142] In this solution, the reconstruction accuracy of video data is improved by mirror projection and search window, and the size of the search window is dynamically adjusted according to the inter-frame motion amplitude to better match the target image block. By decompressing and merging the decompressed video blocks, high-quality decoded video data is finally obtained. It can improve the compression rate and decompression quality of video data and is suitable for various scenarios of video encoding and decoding.

[0143] The video data processing device includes a processor and a memory. The acquisition unit, the block unit, the compression unit, the first merging unit and the sending unit are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions. The modules are all located in the same processor; or, the modules are located in different processors in any combination.

[0144] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problem of low transmission efficiency of coal mine monitoring video data in the prior art can be solved by adjusting kernel parameters.

[0145] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0146] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for processing video data.

[0147] An embodiment of the present invention provides a processor, and the processor is used to run a program, wherein the program executes the method for processing video data when running.

[0148] The embodiment of the present invention provides a device, the device includes a processor, a memory and a program stored in the memory and can be run on the processor, and the processor implements at least the steps of the video data processing method when executing the program. The device in this article can be a server, a PC, a PAD, a mobile phone, etc.

[0149] A computer program product includes a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the video data processing method in each embodiment of the present application are implemented.

[0150] The present application also provides a video data processing system, comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing any of the above-mentioned video data processing methods.

[0151] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0152] 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 in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0153] 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.

[0154] 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 1A function specified in one or more boxes.

[0155] 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 in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0156] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0157] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0158] 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, 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.

[0159] 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.

[0160] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0161] 1) The video data processing method of the present application can decompose the video data of the coal mine into multiple small video blocks, so that when compressing the video blocks, the compression speed is faster, and many small compressed video blocks after compression are merged into an overall compressed video data, so that the video file is smaller and the data is sent faster, which can greatly improve the transmission efficiency.

[0162] 2) The video data processing method of the present application can decompose the video data of the coal mine into multiple small video blocks, so that when compressing the video blocks, the compression speed is faster, and many small compressed video blocks after compression are merged into an overall compressed video data, so that the video file is smaller and the data is sent faster, which can greatly improve the transmission efficiency.

[0163] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for processing video data, characterized in that: include: Obtain video data of coal mines; The video data is processed into blocks to obtain a plurality of video blocks; Compressing all the video blocks to obtain multiple compressed video blocks; Merging all the compressed video blocks to obtain compressed video data; The compressed video data is sent to a target terminal.

2. The method according to claim 1, characterized in that Obtain video data of coal mines, including: Acquiring initial video data of the coal mine; The initial video data is preprocessed to obtain the video data, wherein the preprocessing method includes denoising and / or enhancement.

3. The method according to claim 1, characterized in that The video data is processed into blocks to obtain a plurality of video blocks, including: Obtaining fixed unit information, wherein the fixed unit information is information of a predefined data block size; The video data is processed into blocks according to the fixed unit information to obtain a plurality of the video blocks.

4. The method according to claim 1, characterized in that: The video data is processed into blocks to obtain a plurality of video blocks, including: Using image recognition technology to perform content analysis on the video data to obtain a recognition result, wherein the recognition result at least includes a region type and / or a richness of region details, and the region type includes a dynamic type and / or a static type; According to the recognition result, the video data is divided into blocks to obtain a plurality of the video blocks, wherein the video blocks divided by the dynamic type of video content are larger than the video blocks divided by the static type of video content, and the richness of regional details is positively correlated with the size of the video blocks.

5. The method according to claim 1, characterized in that: All the video blocks are compressed to obtain a plurality of compressed video blocks, including: Calculating the Bhattacharyya coefficients of the Nth video block and the N+1th video block, where N≥1; Determining the importance between the Nth video block and the N+1th video block according to the Bhattacharyya coefficient, wherein the magnitude of the Bhattacharyya coefficient is positively correlated with the importance; Sampling rates are allocated to all the video blocks according to the importance, and the video blocks are compressed according to the sampling rates to obtain a plurality of compressed video blocks, wherein the importance and the sampling rates are positively correlated.

6. The method according to claim 1, characterized in that All the video blocks are compressed to obtain a plurality of compressed video blocks, including: Constructing a video compression model, wherein the video compression model is trained using multiple sets of training data, each set of training data in the multiple sets of training data includes historical video blocks obtained within a historical time period and historical compressed video blocks corresponding to the historical video blocks; The video block is input into the video compression model to obtain the compressed video block corresponding to the video block.

7. The method according to any one of claims 1 to 6, characterized in that: After sending the compressed video data to the target terminal, the method further includes: Performing mirror projection on the image block of the compressed video data to obtain an extended reference frame; Generating a search window in the extended reference frame, wherein the size of the search window is positively correlated with the inter-frame motion amplitude of the compressed video data; In the search window, by means of rasterization traversal, target image blocks that best match the current image block are sequentially extracted to obtain reconstructed video data; Decompressing the reconstructed video data to obtain a plurality of decompressed video blocks; The multiple decompressed video blocks are merged according to their positions during encoding to obtain decoded video data.

8. A video data processing device, characterized in that: include: An acquisition unit, used for acquiring video data of the coal mine; A block division unit, used for performing block processing on the video data to obtain a plurality of video blocks; A compression unit, configured to compress all the video blocks to obtain a plurality of compressed video blocks; A first merging unit, configured to merge all the compressed video blocks to obtain compressed video data; The sending unit is used to send the compressed video data to a target terminal.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for processing video data described in any one of claims 1 to 7 are implemented.

10. A video data processing system, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for processing video data according to any one of claims 1 to 7.

Images