Video security broadcast monitoring method and device based on video genes

Abstract

The invention relates to the field of network audio-visual services, and particularly provides a video security broadcast monitoring method and device based on video genes, and the method comprises the following steps: S1, a monitoring terminal receives a video stream; s2, extracting a key frame to obtain a screenshot; s3, compressing the size of the video and carrying out gray processing; s4, calculating a video gene characteristic value; s5, uploading the characteristic values to a center; s6, comparing the characteristic values; s7, judging a characteristic value; and S8, returning the terminal video. Compared with the prior art, the method can greatly save the normal back-to-source bandwidth and improve the performance. A problem video is quickly pre-positioned through gene comparison, and safety broadcast judgment is started in real time.

Description

Technical Field

[0001] The present invention relates to the field of network audio-visual services, and specifically provides a video broadcast monitoring method and device based on video genes. Background Art

[0002] Current terminal video broadcast monitoring technology generally adopts a completely terminal judgment or completely central judgment method. The completely terminal judgment method generally uses a dedicated server, which is deployed to nodes in various cities and towns. Each of them deploys and configures related services to pull video streams for broadcast monitoring. All processes are handed over to the monitoring terminal for processing, which has high requirements for terminal performance. The terminal equipment is expensive, and the deployment and upgrade are complex and costly. The completely central judgment method requires the city terminal to return the video stream to the center, and then the center makes the judgment. This method requires the complete video stream to be returned, which takes up a lot of return bandwidth. Calculated based on the high-definition video bit rate of 5Mbps, monitoring 200 channels requires 1Gbps bandwidth. Corresponding to the optical and electrical export bandwidth, 1Gbps is expensive and affects other businesses. In addition, the center still needs a large number of servers for parsing and comparison. Summary of the Invention

[0003] The present invention aims to address the above-mentioned deficiencies in the prior art and provides a highly practical video broadcast monitoring method based on video genes.

[0004] A further technical task of the present invention is to provide a video broadcast monitoring device based on video genes that is rationally designed, safe and applicable.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] A video broadcast monitoring method based on video genes has the following steps:

[0007] S1, the monitoring terminal receives the video stream;

[0008] S2, extract key frames to obtain screenshots;

[0009] S3, compress video size and grayscale processing;

[0010] S4, calculating the video gene feature value;

[0011] S5, upload the characteristic value to the center;

[0012] S6, compare feature values;

[0013] S7, determining the characteristic value;

[0014] S8, terminal video transmission.

[0015] Furthermore, in step S1, the monitoring terminal starts to receive video stream data, the video is a multicast stream or an HLS video stream of the HTTP protocol, and after receiving the video stream, the terminal performs decoding and analysis.

[0016] Furthermore, in step S2, the monitoring terminal extracts key frames from the video stream and obtains screenshots, and extracts appropriate key frames from the video for a period of time;

[0017] In step S3, the monitoring terminal further compresses the screenshot, including reducing the video size and processing the RGB color image into a black and white grayscale image.

[0018] Furthermore, in step S4, the gene feature value of the video stream is calculated, and the gene feature value is the feature on the key frame image of the following video;

[0019] In step S5, the calculated characteristic values ​​are uploaded to the central server.

[0020] Furthermore, in step S6, the central monitoring program performs the operations of steps S1 to S5 above, calculates the gene feature value of the central video, and then compares the uploaded feature value with the central gene feature value.

[0021] In step S7, if the eigenvalues ​​are consistent, no processing is performed, and the central city continues to calculate and compare their respective genes; if the eigenvalues ​​are inconsistent, the center initiates a video return instruction to the city terminal.

[0022] Furthermore, in step S8, the center initiates a video return instruction to the city terminal, requiring the cached video of the corresponding time point to be returned to the center. After receiving the instruction, the terminal returns the cached video of the corresponding time point to the center for subsequent judgment operations.

[0023] Furthermore, after transmitting the video back, the center calls the general safe playback algorithm to further calculate the safe playback issues of the abnormal video, determine whether there are flowery screens, black screens, mosaics, and blockiness, and finally, determine the specific safe playback issues and issue a system alarm, while retaining the problematic video for subsequent operations.

[0024] A video broadcast monitoring device based on video genes, comprising: at least one memory and at least one processor;

[0025] The at least one memory is configured to store a machine-readable program;

[0026] The at least one processor is configured to call the machine-readable program to execute a video broadcast monitoring method based on video genes.

[0027] Compared with the prior art, the video gene-based video broadcast monitoring method and device of the present invention have the following outstanding beneficial effects:

[0028] (1) Using the video genetic algorithm, only the characteristic values ​​of the video are calculated and returned, and the return bandwidth only requires one percent of the traditional method.

[0029] (2) The present invention only needs to quickly calculate the video gene feature value. Only when the feature value deviation exceeds the threshold, it is necessary to further determine the broadcast problem, which greatly saves computing resources.

[0030] (3) The calculation speed of the eigenvalue is much faster than the judgment of the broadcast problem. The problem video can be quickly located and then the subsequent complex judgment can be carried out. Moreover, the upgrade, addition of judgment methods and judgment content do not require the upgrade of the terminal's software and hardware. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0032] Figure 1 The present invention is a flowchart of a video broadcast monitoring method based on video genes. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the solutions of the present invention, the present invention will be further described in detail below in conjunction with specific embodiments. Obviously, the embodiments described are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0034] A best embodiment is given below:

[0035] like Figure 1 As shown, a video broadcast monitoring method based on video genes in this embodiment has the following steps:

[0036] S1, the monitoring terminal receives the video stream;

[0037] The monitoring terminal starts to receive video stream data, which can be a multicast stream or an HLS video stream using the HTTP protocol. After receiving the video stream, the terminal decodes and parses it.

[0038] S2, extract key frames to obtain screenshots;

[0039] The monitoring terminal extracts key frames from the video stream and takes screenshots. One or two key frames per second of video are sufficient. Because video genes can be used for similarity comparison, there's no need to generate screenshots for every frame.

[0040] S3, compress video size and grayscale processing;

[0041] This step is optional. During the video size and grayscale processing, the monitoring terminal further compresses the screenshot, including reducing the video size and converting RGB color images into black and white grayscale images. This significantly reduces the video size, speeds up gene generation efficiency, and reduces the size of video genes without affecting the characteristics of the video genes. This reduces the terminal hardware requirements and the bandwidth requirements for network backhaul.

[0042] S4, calculating the video gene feature value;

[0043] Calculate the genetic feature values ​​of the video stream. Genetic feature values ​​are generated by following the features of the video keyframe images, such as corner points, lines, color depth, etc. Algorithms can use surf, binarization, etc.

[0044] S5, upload the characteristic value to the center;

[0045] The calculated eigenvalues ​​are uploaded to the central server. Because the eigenvalues ​​are uploaded to the central server, the bandwidth back to the source is greatly reduced.

[0046] S6, compare feature values;

[0047] The central monitoring program performs the above steps S1 to S5 to calculate the gene feature value of the central video, and then compares the uploaded feature value with the central gene feature value.

[0048] S7, determining the characteristic value;

[0049] If the eigenvalues ​​are consistent, no processing is done and the central city continues to calculate and compare their respective genes.

[0050] If the characteristic values ​​are inconsistent, the center will initiate a video backhaul instruction to the local terminal.

[0051] S8, terminal video transmission;

[0052] The center sends a video return instruction to the local city terminal, requiring it to return the cached video of the corresponding time point to the center.

[0053] After receiving the instruction, the terminal transmits the cached video of the corresponding time point back to the center for subsequent judgment operations.

[0054] After the video is transmitted back, the center calls the general safe playback algorithm to further calculate the safe playback issues of the abnormal video and determine whether there are problems such as flowery screen, black screen, mosaic, blockiness, etc.

[0055] Finally, the specific safe playback problem is determined and a system alarm is issued. At the same time, the problematic video is retained for subsequent operations, such as transferring it to the network quality department for determining network problems.

[0056] The terminal and server use the same algorithm to calculate the video's genetic data. The algorithm first captures keyframe screenshots of the video and then uses them to calculate the video's genetic data. During the calculation, image compression and grayscale binarization can be performed to accelerate the generation of genetic data and reduce its size.

[0057] The central comparison program obtains the real-time video gene data from the monitoring terminal and compares it with the central program's real-time video gene data. This comparison can be performed using algorithms such as the Hamming distance algorithm. Because consistency between the central and user terminal videos is important, the comparison program performs an offset comparison. The central program's current gene value is first compared with the key frame's video gene value from the terminal 30 seconds before and after. Alignment is performed starting with the frame with the highest similarity above the threshold, and the comparison continues with the next frame. Once the specified number of frames has been met, monitoring comparison officially begins.

[0058] When a gene comparison result falls below the threshold, the terminal's recorded video is immediately dispatched and uploaded to the center for actual Anbo monitoring, including black screens, frozen frames, distorted screens, mosaics, etc. Once detected, it is recorded and saved, and a real-time alarm is generated to determine if an Anbo problem has occurred.

[0059] Based on the above method, a video broadcast monitoring device based on video genes in this embodiment includes: at least one memory and at least one processor;

[0060] The at least one memory is configured to store a machine-readable program;

[0061] The at least one processor is configured to call the machine-readable program to execute a video broadcast monitoring method based on video genes.

[0062] The above-mentioned specific implementation methods are only specific cases of the present invention. The patent protection scope of the present invention includes but is not limited to the above-mentioned specific implementation methods. Any technical solutions that conform to the above-mentioned specific implementation methods of the present invention and any appropriate changes or substitutions made thereto by ordinary technicians in the relevant technical field shall fall within the patent protection scope of the present invention.

[0063] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A video broadcast monitoring method based on video gene, characterized in that: The steps are as follows: S1, the monitoring terminal receives the video stream; S2, extract key frames to obtain screenshots; S3, compress video size and grayscale processing; S4, calculating the video gene feature value; S5, upload the characteristic value to the center; S6, compare feature values; S7, determining the characteristic value; S8, terminal video transmission.

2. The video broadcast monitoring method based on video gene according to claim 1, characterized in that: In step S1, the monitoring terminal starts to receive video stream data, which is a multicast stream or an HLS video stream of the HTTP protocol. After receiving the video stream, the terminal decodes and parses it.

3. The video broadcast monitoring method based on video gene according to claim 2 is characterized in that: In step S2, the monitoring terminal extracts key frames from the video stream and obtains screenshots, extracting appropriate key frames from the video for a period of time; In step S3, the monitoring terminal further compresses the screenshot, including reducing the video size and processing the RGB color image into a black and white grayscale image.

4. The video broadcast monitoring method based on video gene according to claim 3 is characterized in that: In step S4, the gene feature value of the video stream is calculated. The gene feature value is the feature on the key frame image of the following video; In step S5, the calculated characteristic values ​​are uploaded to the central server.

5. The video broadcast monitoring method based on video gene according to claim 4 is characterized in that: In step S6, the center monitoring program performs the operations of steps S1 to S5 above, calculates the gene feature value of the center video, and then compares the uploaded feature value with the center gene feature value. In step S7, if the eigenvalues ​​are consistent, no processing is performed, and the central city continues to calculate and compare their respective genes; if the eigenvalues ​​are inconsistent, the center initiates a video return instruction to the city terminal.

6. The video broadcast monitoring method based on video gene according to claim 5, characterized in that: In step S8, the center initiates a video return instruction to the city terminal, requiring the cached video of the corresponding time point to be returned to the center. After receiving the instruction, the terminal returns the cached video of the corresponding time point to the center for subsequent judgment operations.

7. The video broadcast monitoring method based on video gene according to claim 6, characterized in that: After the video is transmitted back, the center calls the general safe playback algorithm to further calculate the safe playback issues of the abnormal video, determine whether there is a flowery screen, black screen, mosaic and blockiness, and finally, determine the specific safe playback problem and issue a system alarm, while retaining the problematic video for subsequent operations.

8. A video broadcast monitoring device based on video gene, characterized in that: include: at least one memory and at least one processor; The at least one memory is configured to store a machine-readable program; The at least one processor is configured to call the machine-readable program to execute the method according to any one of claims 1 to 7.

Images