A method and system for video edge preprocessing and playback

Abstract

The present application provides a method and system for video edge preprocessing and playback. By obtaining the video stream address of the monitoring video source, setting a pulling mechanism according to the video stream address to perform partial pulling processing on the video data, performing protocol conversion on the pulled video data, slicing it according to a preset model to generate an index file, searching for the index file in the video cloud according to the request information, and distributing and transmitting the corresponding video data to the user side. Before the user requests to watch the video, the present application performs partial pulling and preprocessing of the video in advance, and segments and slices the video through intelligent video preprocessing technology, effectively improving the speed during video viewing and playback, from the original loading buffer time of more than 10 seconds to within 3 seconds, improving the user experience; performing protocol conversion on the edge node, avoiding the consumption of bandwidth and server resources when a large number of videos are accessed, and saving the operation cost.

Description

Technical Field

[0001] The present application belongs to the field of video processing technology, and more specifically, relates to a method and system for video edge preprocessing and playback. Background Art

[0002] The digital transformation and smart city initiative require remote monitoring of various businesses, such as farmers' markets, school cafeterias, and hotels supporting the Asian Games. This requires integrating existing video surveillance systems into the monitoring system and playing them on cockpit screens, web applications, and other terminals. Many surveillance devices typically only support RTSP video output, but playback on terminal screens and web applications requires plug-ins, so direct RTSP cannot be supported. This necessitates the introduction of an RTSP to HLS conversion process throughout the entire video surveillance system.

[0003] If RTSP to HLS protocol conversion is centralized on the server, it will significantly increase server bandwidth costs and resource overhead. For large-scale video access, it's often more efficient to use edge nodes for protocol conversion. However, due to issues with the HLS conversion mechanism, it typically takes over 10 seconds from receiving a user's playback request to pulling the video stream from the NVR, starting HLS slicing, and finally enabling user playback, resulting in a poor user experience. Using the RTMP protocol for live streaming, while effectively addressing playback quality issues, consumes significant bandwidth, making it unsuitable for large-scale video promotion.

[0004] In addition, many companies do not have fixed IP addresses, so the server and monitoring equipment cannot communicate directly when accessing video. Video surveillance equipment may belong to different manufacturers and models, making it difficult for the server to interact directly with the equipment. Summary of the Invention

[0005] In view of this, the present application provides a method and system for video edge preprocessing and playback, which pulls and preprocesses the video in advance before the user requests to watch the video, effectively improving the video streaming playback speed and solving the problem of poor user experience.

[0006] The specific technical solutions of this application are as follows:

[0007] This application provides a method for video edge preprocessing and playback, comprising the following steps:

[0008] Collect video data from the surveillance video source, obtain the video stream address of the surveillance video source, set the stream pulling mechanism according to the video stream address, and perform partial stream pulling processing on the video data;

[0009] After protocol conversion of the video data processed by the pull stream, slice it according to a preset model to generate an index file, and perform regular data updates on the index file and upload it to the video cloud;

[0010] Receive a request signal from the user side, search for the index file in the video cloud according to the request information, and distribute and transmit the corresponding video data to the user side.

[0011] Furthermore, the specific method for obtaining the video stream address of the monitoring video source is as follows:

[0012] Deploy an edge video gateway on the enterprise side and directly connect it to the monitoring video source through the intranet;

[0013] Use the edge video gateway to obtain the video stream addresses within the same network segment through automatic scanning technology.

[0014] Furthermore, the specific method for setting the pull stream mechanism according to the video stream address is as follows:

[0015] Analyze the protocol type of the video stream address, and perform partial pull of the video data according to different pull stream mechanisms set according to different protocol types.

[0016] Furthermore, the specific method for setting the pull stream mechanism according to the video stream address is as follows:

[0017] Obtain the RTSP stream address of the monitoring video source, maintain a long-term pull of the video data with the RTSP stream address, and maintain on-demand fetching of the video data with other protocol type stream addresses.

[0018] Furthermore, the specific method for protocol conversion of the video data processed by the pull stream is as follows:

[0019] Automatically perform pre-conversion of the HLS protocol on the video data with the RTSP stream address;

[0020] Perform HLS protocol conversion on the video data with other protocol type stream addresses after on-demand fetching.

[0021] Furthermore, the specific method for slicing and generating an index file according to a preset model is as follows:

[0022] Read in the TS stream, and slice the video data into a series of media segment files of equal time length according to a preset interval time;

[0023] Set index tags according to the identification information of each media segment file and generate an index file.

[0024] Furthermore, the specific method for performing regular data updates on the index file and uploading it to the video cloud is as follows:

[0025] Package and pack the index file according to a preset loading amount, and upload the packaged index file to the video cloud according to a preset update time.

[0026] Furthermore, the specific process of regularly updating and uploading the index file to the video cloud is as follows:

[0027] Arrange the index file according to the recognition information and pre-upload it to each partition of the video cloud;

[0028] Regularly update the index file under each partition at a preset interval and package and encapsulate the index file according to a preset loading amount.

[0029] Furthermore, the specific process of searching for the index file in the video cloud according to the request information is as follows:

[0030] Identify the basic element information and personalized setting information in the request information;

[0031] Match the index tags according to the basic element information and personalized setting information;

[0032] Retrieve the video data in the corresponding index file according to the index tags.

[0033] This application also provides a system for video edge preprocessing and playing, including a memory and a processor. The memory includes a video edge preprocessing and playing program. When the video edge preprocessing and playing program is executed by the processor, the following steps are implemented:

[0034] Collect video data from the monitoring video source, obtain the video stream address of the monitoring video source, and set a pulling mechanism according to the video stream address to perform partial pulling processing on the video data;

[0035] After the protocol conversion of the pulled video data, slice it according to a preset model to generate an index file, and regularly update the data of the index file and upload it to the video cloud;

[0036] Receive a request signal from the using end, search for the index file in the video cloud according to the request information, and distribute and transmit the corresponding video data to the using end.

[0037] Furthermore, the specific process of obtaining the video stream address of the monitoring video source is as follows:

[0038] Deploy an edge video gateway on the enterprise side and directly connect it to the monitoring video source through the intranet;

[0039] Use the edge video gateway to obtain the video stream address within the same network segment through automatic scanning technology.

[0040] Furthermore, the specific process of setting a pulling mechanism according to the video stream address is as follows:

[0041] Analyze the protocol type of the video stream address, and perform partial pulling of video data according to different setting pulling mechanisms for different protocol types.

[0042] Further, setting the pulling mechanism according to the video stream address is specifically as follows:

[0043] Obtain the RTSP stream address of the monitoring video source, maintain long-term pulling of the video data of the RTSP stream address, and maintain on-demand fetching of the video data of the video stream addresses of other protocol types.

[0044] Further, the protocol conversion of the video data after pulling processing is specifically as follows:

[0045] Automatically perform pre-conversion of the HLS protocol on the video data of the RTSP stream address;

[0046] Perform HLS protocol conversion on the video data of the video stream addresses of other protocol types after on-demand fetching.

[0047] Further, slicing and generating an index file according to a preset model is specifically as follows:

[0048] Read in the TS stream, and slice the video data into a series of media segment files of equal time length according to a preset interval time;

[0049] Set index tags according to the identification information of each media segment file and generate an index file.

[0050] Further, the regular data update of the index file and uploading it to the video cloud is specifically as follows:

[0051] Package and pack the index file according to a preset loading amount, and upload the packaged index file to the video cloud according to a preset update time.

[0052] Further, the regular data update of the index file and uploading it to the video cloud is specifically as follows:

[0053] Arrange the index file according to the identification information and pre-upload it to each partition of the video cloud;

[0054] Regularly update the index files under each partition according to a preset interval time and package and pack the index files according to a preset loading amount.

[0055] Further, searching for the index file in the video cloud according to the request information is specifically as follows:

[0056] Identify the basic element information and personalized setting information in the request information;

[0057] Match the index tags according to the basic element information and personalized setting information;

[0058] Retrieve the video data in the corresponding index file according to the index tag.

[0059] In summary, the present application provides a method and system for video edge preprocessing and playing. By obtaining the video stream address of the monitoring video source, setting a pulling mechanism according to the video stream address to perform partial pulling processing on the video data, performing protocol conversion on the pulled video data, slicing it according to a preset model to generate an index file, searching for the index file in the video cloud according to the request information, and distributing and transmitting the corresponding video data to the user side. Before the user requests to watch the video, the present application performs partial pulling and preprocessing of the video in advance, and slices the video through intelligent video preprocessing technology, effectively improving the speed during video viewing and playing, from the original loading buffer time of more than 10 seconds to within 3 seconds, improving the user experience; performing protocol conversion on the edge node, avoiding the consumption of bandwidth and server resources when a large number of videos are accessed, and saving the operating cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0061] Figure 1 It is a flowchart of a method for video edge preprocessing and playing according to the present application.

[0062] Figure 2 It is a schematic diagram of the video edge preprocessing and playing process in the embodiment of the present application.

[0063] Figure 3 It is a block diagram of a system for video edge preprocessing and playing according to the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] In order to make the objectives, features, and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application are clearly and completely described. Obviously, the embodiments described below are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0065] Please refer to Figure 1 , Figure 1 It is a flowchart of a method for video edge preprocessing and playing according to the present application.

[0066] The embodiment of the present application provides a method for video edge preprocessing and playback, characterized by comprising the following steps:

[0067] S102: Collect video data from a surveillance video source, obtain a video stream address of the surveillance video source, and set a stream pulling mechanism according to the video stream address to perform partial stream pulling on the video data;

[0068] S104: After the video data after the stream processing is processed, the protocol is converted and the index file is generated according to the preset model slice, and the index file is regularly updated and uploaded to the video cloud;

[0069] S106: Receive a request signal from the user end, search the index file in the video cloud according to the request information, and distribute and transmit the corresponding video data to the user end.

[0070] It should be noted that the traditional video surveillance system process generally adopts two forms: "on-demand streaming" or "live streaming", and the system uses a level-by-level request. Figure 2 As shown, "on-demand streaming" usually only establishes a complete video link. There is no video stream transmission during the idle period, that is, there is no video stream transmission in channels ①, ②, and ③ in the figure. Therefore, very little bandwidth and server resources are occupied. When a user requests to play a video on the playback terminal, the system will make requests step by step, and only when the video transmission is finally issued will there be video communication. "Live streaming" is the continuous push of live video streams for a long time. Regardless of whether there are users watching the video, the video stream data will occupy the channel, consuming bandwidth and server resources. The advantage is that the buffering time during video playback is short and the experience is good. There is continuous video stream data in channels ①, ②, and ③ in the figure. The embodiments of the present application optimize the entire business process and video pre-processing technology by combining actual business needs, combining the advantages of "on-demand streaming" or "live streaming push", splitting the entire link, and using the edge video gateway to perform uninterrupted HLS slicing in the intranet environment, giving full play to the advantages of "live streaming push", while continuing the "on-demand streaming" model in the extranet environment, providing a better user experience at a low cost.

[0071] The streaming mechanism in S102 can be set according to the type of video stream address. For example, a continuous streaming mechanism is adopted for video stream addresses with high demand, while an on-demand streaming mechanism is adopted for video stream addresses with low demand. The protocol conversion in S104 can adopt conventional methods in this field, and the preset slicing model can be determined based on information such as the size of the video data, feature node positions or traffic. Uploading index files to the cloud regularly can cache data in advance to prevent excessive waiting time. In S106, the user can manually initiate a request message, and then search and divert traffic in the cloud according to the set content of the request message, thereby further improving the video playback speed.

[0072] According to an embodiment of the present application, obtaining the video stream address of the monitoring video source specifically includes:

[0073] Deploy an edge video gateway on the enterprise side and directly connect it to the monitoring video source through the intranet;

[0074] Use the edge video gateway to obtain the video stream address within the same network segment through automatic scanning technology.

[0075] It should be noted that the enterprise side refers to the source or processor of video data. The edge video gateway can be connected to various video monitoring devices (NVR or IPC), supports multi-video stream access compatibility, can effectively compatible with the rtsp stream access of various types of devices, solves the problem of reusing and accessing various original cameras, and there is no need to reinstall monitoring devices and rewire, with obvious advantages. The automatic scanning technology can be a conventional method for collecting video stream addresses in this field.

[0076] According to an embodiment of the present application, setting the video stream pulling mechanism according to the video stream address specifically includes:

[0077] Analyze the protocol type of the video stream address, and perform partial pulling of video data according to different setting video stream pulling mechanisms for different protocol types.

[0078] It should be noted that the protocol types of the video stream address may include RTMP, RTSP, HTTP, etc. Set the video stream pulling mechanism according to the demand degree and processing difficulty of the protocol type, perform continuous preprocessing on a part of the video data, which can improve the user experience without reducing the memory consumption of video processing; perform processing shunting on another part of the video data with low demand or high processing consumption when requested. As Figure 2 shown in the figure, part ① in the figure will continuously maintain the video stream transmission. This part retains the advantages of real-time processing of "live push stream" and saves the time-consuming of "pulling stream on demand" each time because it is an intranet and the consumption of bandwidth cost is not large.

[0079] According to an embodiment of the present application, setting the video stream pulling mechanism according to the video stream address specifically includes:

[0080] Obtain the RTSP stream address of the monitoring video source, maintain long-term pulling of the video data of the RTSP stream address, and maintain on-demand pulling of the video data of other protocol type stream addresses.

[0081] It should be noted that selecting the video data of the RTSP stream address for continuous pulling and slicing can maximize the video preprocessing efficiency and video playback speed. The video stream address for long-term pulling can also be set to other protocol types, which can play a complementary role in different degrees and scenarios. On-demand pulling also goes through the processes of video stream address parsing, protocol conversion, and slicing shunting.

[0082] According to the embodiments of the present application, the protocol conversion of the video data after the pull stream processing is specifically as follows:

[0083] Automatically perform pre - conversion of the HLS protocol on the video data of the RTSP stream address;

[0084] For the video data of other protocol type stream addresses, perform HLS protocol conversion after fetching the stream on demand.

[0085] It should be noted that since the time consumption of converting from RTSP to the HLS protocol after pulling the video stream at the edge gateway is one of the major factors affecting the user experience. The embodiments of the present application pre - perform protocol conversion on the video data of the RTSP stream address, which can greatly improve the video playback speed and will not occupy too much processing memory. Both the protocol conversion and the parsing of the video stream address are long - term and continuous. Among them, the conversion method of the HLS protocol can adopt the conventional methods in the field.

[0086] According to the embodiments of the present application, the specific process of generating an index file by slicing according to a preset model is as follows:

[0087] Read the TS stream, and slice the video data into a series of media shard files of equal time length according to a preset interval time;

[0088] Set index tags according to the identification information of each media shard file and generate an index file.

[0089] It should be noted that the TS stream is a file format, which can display video data as media shard files. Slicing the video data according to a preset interval time means slicing the video data into equal - duration slices, and the equal time length can be set to 10 seconds. In addition, the preset interval time can be reasonably adjusted according to the total duration or total capacity information of the video data, and the purpose is to reduce the video shunt waiting time.

[0090] According to the embodiments of the present application, the specific process of regularly updating the data of the index file and uploading it to the video cloud is as follows:

[0091] Package and pack the index file according to a preset loading amount, and upload the packaged index file to the video cloud according to a preset update time.

[0092] It should be noted that in a traditional video processing system, at least 3 fragmented videos must be loaded before the current fragment can be started for playback. The fragmented duration of the HLS standard is 10s. Loading 3 fragments means that the standard latency reaches 30s, which is unacceptable in a normal live broadcast scenario. In the embodiment of the present application, through the edge video gateway, conversion slicing can be pre-performed before a user requests playback, and the file can be continuously updated. When a user requests to access and play a video, a quick response can be made without waiting for the time-consuming of multiple video slices, greatly reducing the latency. It can not only utilize the wide adaptability advantage of the HLS protocol for various terminal playbacks, but also effectively solve the latency drawback of HLS. Among them, the preset loading amount can be determined according to the fragmented duration of the index file, and the video data is directly uploaded to the cloud and sent to the terminal after being encapsulated and packed. The preset update time can be determined according to the total duration of the video data, and the data update is mainly used for error checking, repair, arrangement, and re-encapsulation of the video data.

[0093] According to the embodiment of the present application, the specific operation of regularly updating and uploading the index file to the video cloud is as follows:

[0094] Arrange the index file according to the identification information and pre-upload it to each partition of the video cloud;

[0095] Regularly update the index file under each partition at a preset interval time and encapsulate and pack the index file according to the preset loading amount.

[0096] It should be noted that the index file can be partitioned and cached in the video cloud according to feature nodes such as the recognized content title, picture switching, sound change, and behavior activation. Then, the video data of each partition is regularly encapsulated and packed, which can more accurately capture the video playback nodes, make the video playback continuous, and enhance the user experience. Among them, the preset interval time and the preset loading amount can both be determined according to the duration of the video data in the video cloud partition.

[0097] According to the embodiment of the present application, the specific operation of searching for the index file in the video cloud according to the request information is as follows:

[0098] Identify the basic element information and personalized setting information in the request information;

[0099] Match the index label according to the basic element information and the personalized setting information;

[0100] Retrieve the video data in the corresponding index file according to the index label.

[0101] It should be noted that the basic element information includes system setting information such as video source, classification, theme, video length, and size, and the personalized setting information includes system data information such as historical video views, historical comment counts, and historical repost counts. By matching the request information set by the user with the index tags of the video data in the video cloud, the corresponding video data is retrieved, avoiding the time loss of temporarily screening video information. As Figure 2 shown, in the embodiment of the present application, ②③ use the Internet and continue to maintain the "fetching stream on demand" mode, and the video stream data is only distributed and transmitted when the user requests to watch, also having the advantage of saving bandwidth.

[0102] In another embodiment of the present application, the video stream addresses within the same network segment are obtained through an automatic scanning technology, specifically:

[0103] Collect the address information of the monitoring video source and the internal encoder;

[0104] Obtain the video stream addresses within the same network segment by analyzing the address information of the monitoring video source and the internal encoder.

[0105] It should be noted that the edge video gateway in the embodiment of the present application has a video stream scanning function. By collecting the address information of the monitoring video source and the internal encoder to obtain the video stream addresses, it can automatically scan all the video stream addresses within the same local area network, facilitating video access and solving the problem of finding video addresses manually.

[0106] In another embodiment of the present application, slicing according to a preset model to generate an index file is specifically:

[0107] Read in the TS stream and identify the mutation time points of the picture pixels, sound frequencies, and traffic hotspots in the video data;

[0108] Set the splitting breakpoints of the video data according to the mutation time points and slice the video data to generate an index file.

[0109] It should be noted that by determining the theme content demarcation points or behavior start demarcation points of the video according to the mutation time points of features such as picture pixels, sound frequencies, and traffic hotspots in the video data, and setting video slice nodes according to these demarcation points, the user interest points can be captured more accurately, so as to classify and manage the videos and fetch the streams in batches, improving the user viewing fluency and comfort.

[0110] In another embodiment of the present application, it further includes:

[0111] Obtain the scene setting information from the usage end;

[0112] Perform interval frame extraction on the video data according to the scene setting information, and perform intelligent analysis on the extracted frame data according to the scene setting information;

[0113] Mark the abnormal results as violation information and push it to the user side.

[0114] It should be noted that this application embodiment provides a frame extraction screenshot and intelligent analysis function. The scenario settings can include farmers' markets, office places, traffic roads, etc. Users can input scenarios, interval times, recognition ranges, and recognition objects in the scenario setting information. For example, set various different time intervals such as 5 seconds, 5 minutes, and 30 minutes at the user side for frame extraction, and push the pictures to the AI analysis unit for intelligent analysis according to the scenarios, recognition ranges, and recognition objects in the business scenario, initially analyzing whether there are people in the pictures and whether there are violations (such as people not wearing safety helmets, work clothes, masks, etc.). Finally, push the analysis results of not recognizing the expected recognition objects or recognizing abnormal recognition objects to the user side.

[0115] Please refer to Figure 3 , Figure 3 which is a system block diagram for video edge preprocessing and playing of this application.

[0116] This application embodiment also provides a system for video edge preprocessing and playing, including a memory 21 and a processor 22. The memory 21 includes a video edge preprocessing and playing program. When the video edge preprocessing and playing program is executed by the processor 22, the following steps are implemented:

[0117] Collect video data from the monitoring video source, obtain the video stream address of the monitoring video source, and perform partial pulling processing on the video data according to the video stream address by setting a pulling mechanism.

[0118] After performing protocol conversion on the pulled video data, slice it according to a preset model to generate an index file, and perform regular data update on the index file and upload it to the video cloud.

[0119] Receive a request signal from the user side, search for the index file in the video cloud according to the request information, and distribute and transmit the corresponding video data to the user side.

[0120] According to this application embodiment, specifically obtaining the video stream address of the monitoring video source is as follows:

[0121] Deploy the edge video gateway on the enterprise side and directly connect it to the monitoring video source through the intranet.

[0122] Use the edge video gateway to obtain the video stream addresses within the same network segment through automatic scanning technology.

[0123] According to this application embodiment, specifically setting the pulling mechanism according to the video stream address is as follows:

[0124] Analyze the protocol type of the video stream address, and perform partial pulling of video data according to different preset pulling mechanisms based on different protocol types.

[0125] According to the embodiments of the present application, setting the pulling mechanism according to the video stream address is specifically as follows:

[0126] Obtain the RTSP stream address of the monitoring video source, maintain long-term pulling of the video data of the RTSP stream address, and maintain on-demand fetching of the video data of the video stream addresses of other protocol types.

[0127] According to the embodiments of the present application, performing protocol conversion on the video data after pulling processing is specifically as follows:

[0128] Automatically perform pre-conversion of the HLS protocol on the video data of the RTSP stream address;

[0129] Perform HLS protocol conversion on the video data of the video stream addresses of other protocol types after on-demand fetching.

[0130] According to the embodiments of the present application, slicing and generating an index file according to a preset model is specifically as follows:

[0131] Read in the TS stream, and slice the video data into a series of media shard files of equal time length according to a preset interval time;

[0132] Set index tags according to the identification information of each media shard file and generate an index file.

[0133] According to the embodiments of the present application, performing regular data update on the index file and uploading it to the video cloud is specifically as follows:

[0134] Package and pack the index file according to a preset loading amount, and upload the packaged index file to the video cloud according to a preset update time.

[0135] According to the embodiments of the present application, performing regular data update on the index file and uploading it to the video cloud is specifically as follows:

[0136] Arrange the index file according to the identification information and pre-upload it to each partition of the video cloud;

[0137] Regularly update the index files under each partition according to a preset interval time and package and pack the index files according to a preset loading amount.

[0138] According to the embodiments of the present application, searching for index files in the video cloud according to the request information is specifically as follows:

[0139] Identify the basic element information and personalized setting information in the request information;

[0140] Match the index tags according to the basic element information and personalized setting information;

[0141] Retrieve the video data in the corresponding index file according to the index tag.

[0142] In another embodiment of the present application, the specific method for obtaining the video stream addresses within the same network segment through the automatic scanning technology is as follows:

[0143] Collect the address information of the monitoring video source and the internal encoder;

[0144] Obtain the video stream addresses within the same network segment by analyzing the address information of the monitoring video source and the internal encoder.

[0145] In another embodiment of the present application, the specific method for slicing and generating the index file according to the preset model is as follows:

[0146] Read in the TS stream and identify the mutation time points of the picture pixels, sound frequencies, and traffic hotspots in the video data;

[0147] Set the split breakpoints of the video data according to the mutation time points and slice the video data to generate the index file.

[0148] In another embodiment of the present application, it further includes:

[0149] Obtain the scene setting information from the user side;

[0150] Perform interval frame extraction on the video data according to the scene setting information, and perform intelligent analysis on the extracted frame data according to the scene setting information;

[0151] Mark the abnormal results as violation information and push them to the user side.

[0152] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical, or other forms.

[0153] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units; they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0154] In addition, each functional unit in the embodiments of the present invention may be all integrated into one processing unit, or each unit may be separately taken as one unit, or two or more units may be integrated into one unit; the above-mentioned integrated unit may be implemented in the form of hardware, or in the form of a hardware plus software functional unit.

[0155] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: removable storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks or optical disks and other various media that can store program codes.

[0156] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the foregoing storage medium includes: removable storage devices, ROM, RAM, magnetic disks or optical disks and other various media that can store program codes.

[0157] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for video edge preprocessing and playback, characterized in that It includes the following steps: Collect video data from the monitoring video source, obtain the video stream address of the monitoring video source, and perform partial pulling of the video data according to the video stream address by setting a pulling mechanism. After the protocol conversion of the pulled video data, slice it according to a preset model to generate an index file, and perform regular data update on the index file and upload it to the video cloud. Receive the request signal from the user side, search for the index file in the video cloud according to the request information, and distribute and transmit the corresponding video data to the user side. Setting the pulling mechanism according to the video stream address specifically is: Analyze the protocol type of the video stream address, and perform partial pulling of the video data according to different setting pulling mechanisms according to different protocol types. Performing regular data update on the index file and uploading it to the video cloud specifically is: Package and pack the index file according to a preset loading amount, and upload the packaged index file to the video cloud according to a preset update time. Performing regular data update on the index file and uploading it to the video cloud specifically is: Arrange the index file according to the recognition information and pre-upload it to each partition of the video cloud. Perform regular update on the index file under each partition according to a preset interval time and package and pack the index file according to a preset loading amount. Obtaining the video stream address in the same network segment through automatic scanning technology specifically is: Collect the address information of the monitoring video source and the internal encoder. Obtain the video stream address in the same network segment by analyzing the address information of the monitoring video source and the internal encoder. It also includes: Obtain the scene setting information from the user side. Perform interval frame extraction on the video data according to the scene setting information, and perform intelligent analysis on the extracted frame data according to the scene setting information. Mark the abnormal result as a violation message and push it to the user side.

2. The method for video edge preprocessing and playback according to claim 1, wherein Obtaining the video stream address of the monitoring video source specifically is: Deploy the edge video gateway on the enterprise side and connect it directly to the monitoring video source through the intranet. Use the edge video gateway to obtain the video stream address in the same network segment through automatic scanning technology.

3. The method for video edge preprocessing and playing according to claim 1, wherein Setting the pulling mechanism according to the video stream address specifically is: Obtain the RTSP stream address of the monitoring video source, maintain long-term pulling of the video data of the RTSP stream address, and maintain on-demand fetching of the video data of other protocol type stream addresses.

4. The method for video edge preprocessing and playing according to claim 1, wherein Performing protocol conversion on the pulled video data specifically is: Automatically perform pre-conversion of the HLS protocol on the video data of the RTSP stream address. Perform HLS protocol conversion on the video data of other protocol type stream addresses after on-demand fetching.

5. The method for video edge preprocessing and playback according to claim 1, wherein Slicing and generating an index file according to a preset model specifically is: Read the TS stream, and slice the video data into a series of media shard files of equal time length according to a preset interval time. Set index tags according to the recognition information of each media shard file and generate an index file.

6. The method for video edge preprocessing and playback according to claim 1, wherein Searching for the index file in the video cloud according to the request information specifically is: Identify the basic element information and personalized setting information in the request information. Match the index tags according to the basic element information and personalized setting information. Retrieve the video data in the corresponding index file according to the index tags.

7. A system for video edge preprocessing and playback, characterized in that It includes a memory and a processor. The memory includes a video edge preprocessing and playback program. When the video edge preprocessing and playback program is executed by the processor, the steps of the video edge preprocessing and playback method according to any one of claims 1 to 6 are implemented.

Images