Multi-computing-power-board edge computing device for public security business
By using a multi-computing-board edge computing device, the problems of high cost and low efficiency in video data stream processing in public security operations have been solved, enabling efficient and flexible video data stream processing and real-time monitoring, which is suitable for different network environments.
Patent Information
- Application Number
- CN202422137687.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2025-11-04
- Estimated Expiration
- 2034-08-30
AI Technical Summary
In existing public security operations, video data stream processing is costly and inefficient, making it impossible to achieve real-time and spatial-range target tracking and timely warning of abnormal target behavior.
It adopts a multi-computing board edge computing device, including a switching chip and an edge computing module, to realize the parsing and processing of video data streams through gigabit/10-gigabit broadband communication, supports network address translation, integrates an embedded neural network chip, supports hot-swapping and status display, and adapts to different network environments.
It improves the efficiency and flexibility of video data stream processing, reduces processing costs, and enables real-time monitoring and timely early warning.
Smart Images

Figure CN223514947U_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The utility model relates to public security business data processing technical field especially relates to a multi -computing board edge computing device for public security business. BACKGROUND
[0002] In public security business, image detection technology has gradually surpassed criminal investigation, technical detection or network detection means. The processing of video data stream in image detection technology is crucial.
[0003] In the implementation of the utility model, it is found that the prior art has the following defects: at present, most staff use image detection technology, and can only rely on traditional technical means of manual video review and video tracking, cannot search target trend in all time and space, and cannot timely warn target abnormal behavior. However, the demand for video image intelligent construction is increasing day by day, but according to the current national public security image detection status, affected by objective conditions such as many old video image equipment, long update cycle or rapid technology development, leading to high cost and low efficiency of public security business video data stream processing. UTILITY MODEL CONTENT
[0004] The utility model provides a multi -computing board edge computing device for public security business to realize the efficiency and flexibility of improving video data stream processing.
[0005] According to an aspect of the utility model, a multi -computing board edge computing device for public security business is provided, wherein the multi -computing board edge computing device includes: switching chip and at least one edge computing module, wherein each edge computing module includes multiple computing boards;
[0006] The switching chip and each computing board communicate through gigabit broadband, for receiving each video data stream, and each video data stream is sent to each computing board;
[0007] Wherein, each video data stream is collected by each camera in the public security business scene respectively, and each video data stream collected is sent to the switching chip, and each camera communicates with the switching chip through gigabit broadband respectively;
[0008] The edge computing module is used to analyze and process each video data stream received through each computing board, and obtain video data stream processing result.
[0009] Further, it further includes: the switching chip is support network address conversion switching chip;Wherein, the switching chip and each computing board corresponding to the edge computing module include one or more transformers.
[0010] Further, each video data stream in each camera is transmitted to the network address translation support switch chip through a first network segment; the network address translation support switch chip performs network address translation processing on the first network segment and transmits to each computing power board through a second network segment obtained.
[0011] Further, each camera is configured to collect one video data stream at the same time.
[0012] Further, the multi-computing power board edge computing device further comprises a main control board; wherein the main control board is configured to receive data interaction service instructions and communicate with the network address translation support switch chip through a gigabit broadband.
[0013] Further, the types of the edge computing module include a first type of edge computing module and a second type of edge computing module; wherein the first type of edge computing module comprises eight computing power boards and one main control board, and is configured to perform edge computing processing on 280 video data streams in a gigabit broadband network environment; the second type of edge computing module comprises ten groups of the first type of edge computing module integrated by eight computing power boards and one main control board, and is configured to perform edge computing processing on 2800 video data streams in a terabit broadband network environment.
[0014] Further, each computing power board supports hot plugging.
[0015] Further, each computing power board corresponds to one embedded neural network chip.
[0016] Further, the multi-computing power board edge computing device further comprises an access computing power board quantity statistical module and a computing power board access video data stream statistical module; wherein the access computing power board quantity statistical module is configured to statistically process the quantity of currently accessed computing power boards; and the computing power board access video data stream statistical module is configured to statistically process the video data streams accessed by each computing power board respectively.
[0017] Further, the multi-computing power board edge computing device further comprises a computing power board state display module; wherein the computing power board state display module comprises a computing power board normal state display unit, a computing power board idle state display unit and a computing power board non-insertion state display unit; the computing power board normal state display unit is connected with a green indicator light, and is configured to display the computing power board in a normal state through the green indicator light; the computing power board idle state display unit is connected with a yellow indicator light, and is configured to display the computing power board in an idle state through the yellow indicator light; and the computing power board non-insertion state display unit is connected with a red indicator light, and is configured to display the computing power board in a non-insertion state through the red indicator light.
[0018] The technical scheme of the embodiment of the utility model discloses a multi-computing power board edge computing device, comprising a switching chip and at least one edge computing module, wherein each edge computing module comprises a plurality of computing power boards, the switching chip and each computing power board communicate through a gigabit broadband, for receiving each video data stream and sending each video data stream to each computing power board, wherein each video data stream is collected by each camera in a public security service scene and sent to the switching chip, each camera communicates with the switching chip through a gigabit broadband, the edge computing module is used for analyzing and processing each received video data stream through each computing power board to obtain a video data stream processing result.
[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the utility model, nor is it used to limit the scope of the utility model. Other features of the utility model will become readily apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS
[0020] In order to more clearly illustrate the technical scheme in the embodiments of the utility model, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the utility model, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.
[0021] Figure 1 It is a structure schematic view of a multi-computing power board edge computing device for public security services according to the embodiment of the utility model;
[0022] Figure 2 It is a structure schematic view of a first type of edge computing module in the device according to the embodiment of the utility model;
[0023] Figure 3 It is a structure schematic view of a second type of edge computing module in the device according to the embodiment of the utility model. DETAILED DESCRIPTION
[0024] In order to make the person skilled in the art better understand the technical scheme of the present application, the technical scheme in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the person skilled in the art without creative labor should belong to the scope of protection of the present application.
[0025] It should be noted that the terms "target", "current" and the like in the specification and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.
[0026] Figure 1 It is a structure schematic view of a multi-computing power board edge computing device for public security business provided according to an embodiment of the present application. The multi-computing power board edge computing device 100 comprises a switching chip 110 and at least one edge computing module 120, wherein each edge computing module 120 comprises a plurality of computing power boards 121.
[0027] The switching chip 110 and each computing power board 121 corresponding to the edge computing module 120 communicate through gigabit broadband, for receiving each video data stream and sending each video data stream to each computing power board 121.
[0028] Each video data stream is collected by each camera in a public security business scene, and each collected video data stream is sent to the switching chip 110. Each camera communicates with the switching chip 110 through gigabit broadband.
[0029] The edge computing module 120 is configured to analyze and process each received video data stream through each computing power board 121 to obtain a video data stream processing result.
[0030] Optionally, each camera is configured to collect one video data stream at the same time.
[0031] The cameras can be deployed in various public security scenarios to acquire real-time video data streams for each scenario. Each camera collects one video data stream simultaneously and processes that stream. For example, assuming N cameras can collect data streams concurrently, N video data streams can be acquired.
[0032] Optionally, the switching chip is a network address translation (NAT) switching chip; wherein, each computing board corresponding to the switching chip and the edge computing module includes one or more transformers.
[0033] Optionally, each video data stream from each of the cameras is transmitted to the network address translation (NAT) switching chip via a first network segment; the NAT switching chip performs NAT processing on the first network segment and transmits it to each of the computing boards via the resulting second network segment.
[0034] The switching chip is a network address translation (NAT) chip, which can convert the network segments of the input and output video data streams. For example, assuming the first network segment of the input video data stream is 192.100.xxx.01 and the processed target network segment (i.e., the second network segment) is 10.11.xxx.11, the NAT chip is needed to convert the network address. This prevents the internal network address from being exposed to the outside world and ensures that the input and output of the video data stream are not in the same network segment, thus guaranteeing data security and confidentiality.
[0035] Optionally, the multi-computing-board edge computing device further includes a main control board; wherein the main control board is used to receive data interaction service instructions and communicate with the network address translation switching chip via gigabit broadband.
[0036] In this embodiment, the main control board can also receive data interaction service instructions, perform data processing according to the received data interaction service instructions, and communicate with the network address translation switching chip and various computing boards through the network address translation switching chip.
[0037] Figure 2 This is a schematic diagram of the structure of the first type of edge computing module in the device provided according to the embodiments of the present utility model. Figure 3 This is a schematic diagram of the structure of a second type of edge computing module in the device provided according to an embodiment of the present invention. Figure 3The edge computing module includes 10 groups of first-type edge computing modules, and communicates with a total master control board and a backup total master control board through an exchange chip, can receive control instructions of the total master control board, and controls the second-type edge computing module through the control instructions. The total master control board and the backup total master control board are redundant to each other, and when the total master control board or the backup total master control board fails, the system automatically switches to the other total master control board.
[0038] Optionally, the types of the edge computing module include a first-type edge computing module and a second-type edge computing module; the first-type edge computing module includes eight computing power boards and one master control board, and is used for edge computing processing of 280 video data streams in a gigabit broadband network environment; the second-type edge computing module includes 10 groups of first-type edge computing modules integrated by eight computing power boards and one master control board, and is used for edge computing processing of 2800 video data streams in a gigabit broadband network environment.
[0039] In the embodiment, generally, each computing power board supports processing of 32 video data streams at most, and for eight computing power boards, processing of 256 video data streams at most can be supported. In addition, 24 video data streams are shared by the master control board, so that the first-type edge computing module can perform edge computing processing of 280 video data streams at most and is applied to a gigabit broadband network environment.
[0040] In addition, the second-type edge computing module is integrated by 10 first-type edge computing modules, each of which can perform edge computing processing of 280 video data streams at most, so that the second-type edge computing module can perform edge computing processing of 2800 video data streams at most.
[0041] Correspondingly, the first-type edge computing module is suitable for a gigabit broadband network environment and performs edge computing processing of 280 video data streams, and the specific application scenario is a community, a college, a large supermarket, or a small scene in a park. The second-type edge computing module is suitable for a gigabit broadband network environment and performs edge computing processing of 2800 video data streams, and the specific application scenario can be a large-scale urban use scenario.
[0042] Both the computing power board including the first-type edge computing module and the computing power board including the second-type edge computing module support hybrid cluster deployment and load balancing, and adapt to arbitrary horizontal expansion and disaster recovery backup.
[0043] Optionally, each computing power board corresponds to an embedded neural network chip.
[0044] In the embodiment, each computing power board can correspond to an embedded neural network chip. For the first type of edge computing module, the size of the module is small, which can be set to 20cmx20cmx10cm, and the average power consumption is as low as 50 watts.
[0045] Optionally, each computing power board supports hot plugging.
[0046] Among them, since each computing power board has the function of supporting hot plugging, different numbers of computing power boards can be set according to actual needs.
[0047] In the embodiment, different numbers of computing power boards can be accessed according to the number of cameras, and since the maximum number of video data streams processed by each computing power board is different, the computing power boards can be more flexibly increased or reduced, improving the flexibility of video data stream processing and better and more reasonably using the computing power boards.
[0048] For a specific example, assuming that in the current scenario, 1000 cameras are included, and 1000 video data streams are collected respectively. Since the first type of edge computing module can perform edge computing processing on a maximum of 280 video data streams, the second type of edge computing module is needed to process 1000 video data streams.
[0049] Specifically, since the second type of edge computing module includes 10 first type of edge computing modules, 4 first type of edge computing modules are needed for processing operation. Assuming that first type of edge computing module 1, first type of edge computing module 2, first type of edge computing module 3 and first type of edge computing module 4 are used respectively; the remaining 6 first type of edge computing modules are in idle state.
[0050] Specifically, first type of edge computing module 1 is responsible for processing the first 280 video data streams; first type of edge computing module 2 is responsible for processing the 281st to 560th video data streams; first type of edge computing module 3 is responsible for processing the 561st to 840th video data streams; and first type of edge computing module 4 is responsible for processing the 841st to 1000th video data streams.
[0051] For the first type of edge computing module 4, it includes 160 video data streams, and since each computing power board in the first type of edge computing module 4 (including computing power board 1, computing power board 2, computing power board 3, computing power board 4, computing power board 5, computing power board 6, computing power board 7, computing power board 8 and main control board) can process a maximum of 32 video data streams, computing power board 1, computing power board 2, computing power board 3, computing power board 4 and computing power board 5 each share 32 video data streams, and computing power board 6, computing power board 7, computing power board 8 and main control board are in idle state.
[0052] In the embodiment, the multi-computing power board edge computing device can further include: an access computing power board quantity statistical module, a computing power board access video data stream statistical module, and a computing power board state display module.
[0053] Optionally, the multi-computing power board edge computing device further includes: an access computing power board quantity statistical module and a computing power board access video data stream statistical module; the access computing power board quantity statistical module is configured to perform statistical processing on the quantity of currently accessed computing power boards; and the computing power board access video data stream statistical module is configured to perform statistical processing on video data streams respectively accessed by each computing power board.
[0054] As a continuation of the foregoing example, 1000 cameras can be accessed by the computing power board access video data stream statistical module, that is, 1000 video data streams are accessed. The quantity of accessed computing power boards can be counted by the access computing power board quantity statistical module. Specifically, 8 computing power boards are accessed by the first type of edge computing module 1, 8 computing power boards are accessed by the first type of edge computing module 2, 8 computing power boards are accessed by the first type of edge computing module 3, and 5 computing power boards are accessed by the first type of edge computing module 4. Therefore, 29 computing power boards are accessed.
[0055] Optionally, the multi-computing power board edge computing device further includes: a computing power board state display module; the computing power board state display module includes a computing power board normal state display unit, a computing power board idle state display unit, and a computing power board uninserted state display unit; the computing power board normal state display unit is connected to a green indicator light, and is configured to display a computing power board in a normal state by the green indicator light; the computing power board idle state display unit is connected to a yellow indicator light, and is configured to display a computing power board in an idle state by the yellow indicator light; and the computing power board uninserted state display unit is connected to a red indicator light, and is configured to display a computing power board in an uninserted state by the red indicator light.
[0056] As a continuation of the foregoing example, for the first type of edge computing module 4, computing power board 1, computing power board 2, computing power board 3, computing power board 4, and computing power board 5 are all in a computing power board normal state, and the state of the computing power boards can be displayed by the green indicator light corresponding to the computing power board normal state display unit.
[0057] Suppose that for computing power board 1, due to a fault, computing power board 1 loses connection, that is, the current computing power board 1 is in a computing power board uninserted state, and the state of the computing power board can be displayed by the red indicator light corresponding to the computing power board uninserted state display unit.
[0058] Further, when the computing power board 1 fails, it is necessary to transfer the 32-way video data stream processed by the computing power board 1 to other computing power boards, but since the computing power boards 2, 3, 4 and 5 are in a full load state, and the computing power boards 6, 7, 8 and the main control board are in an idle state, the 32-way video data stream in the computing power board 1 can be transferred to the computing power board 6, and the state of the computing power board 6 is updated.
[0059] Correspondingly, it is assumed that the computing power board 1 can be normally connected after 10s, and the computing power board 1 is in an idle state at this time, and the idle state of the computing power board is displayed by the yellow indicator lamp corresponding to the computing power board idle state display unit.
[0060] In addition, the multi-computing power board edge computing device further comprises a camera state display module, wherein the camera state display module is configured to display the online state, offline state and unassigned state of each camera.
[0061] In the embodiment, the state of each camera can also be monitored, and the online state, offline state or unassigned state of the camera is monitored in real time. The camera in the offline state can be timely alarmed to ensure the normal operation of the camera. The video data stream corresponding to the camera in the unassigned state can be timely assigned, so that the efficiency and real-time performance of the video data stream processing can be improved.
[0062] The technical scheme of the embodiment of the utility model discloses a multi-computing power board edge computing device comprising a switching chip and at least one edge computing module, wherein each edge computing module comprises a plurality of computing power boards, the switching chip and each computing power board corresponding to the edge computing module communicate through a gigabit broadband, are used for receiving each video data stream and sending each video data stream to each computing power board, wherein each video data stream is collected by each camera in a public security business scene and sent to the switching chip, and each camera communicates with the switching chip through a gigabit broadband, and the edge computing module is used for analyzing and processing each received video data stream through each computing power board to obtain a video data stream processing result.
[0063] The above specific embodiments do not constitute a limitation on the protection scope of the utility model. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principles of the utility model should be included in the protection scope of the utility model.
Claims
1. A multi-computing-power edge computing device for public security operations, characterized in that, The multi-computing-board edge computing device includes: a switching chip and at least one edge computing module, wherein each edge computing module includes multiple computing boards; The switching chip communicates with each computing board corresponding to the edge computing module via gigabit broadband to receive each video data stream and send each video data stream to each computing board. Each of the video data streams is collected by a camera in a public security business scenario, and the collected video data streams are sent to the switching chip. Each camera communicates with the switching chip via 10 Gigabit broadband. The types of edge computing modules include: a first type of edge computing module and a second type of edge computing module; The first type of edge computing module includes 8 computing boards and 1 main control board, which are used to perform edge computing processing on 280 video data streams in a gigabit broadband network environment. The second type of edge computing module includes 10 sets of first type edge computing modules integrated by 8 computing boards and 1 main control board, which are used to perform edge computing processing on 2,800 video data streams in a 10 Gigabit broadband network environment; This also includes: each of the aforementioned computing boards supports hot-swapping; This also includes: each computing board corresponds to an embedded neural network chip.
2. The apparatus according to claim 1, characterized in that, Also includes: The switching chip is a network address translation switching chip; Each of the switching chip and each computing board corresponding to the edge computing module includes one or more transformers.
3. The apparatus according to claim 2, characterized in that, Also includes: Each video data stream from each of the cameras is transmitted to the network address translation switching chip via the first network segment; The network address translation switching chip performs network address translation processing on the first network segment and transmits the resulting second network segment to each of the computing boards.
4. The apparatus according to claim 3, characterized in that, Also includes: Each of the cameras is used to acquire one video data stream at the same time.
5. The apparatus according to claim 4, characterized in that, The multi-computing board edge computing device also includes: a main control board; The main control board is used to receive data interaction service instructions and communicate with the network address translation switching chip via gigabit broadband.
6. The apparatus according to claim 5, characterized in that, The multi-computing board edge computing device further includes: a module for counting the number of connected computing boards and a module for counting the video data stream connected to the computing boards; The connected computing board number statistics module is used to perform statistical processing on the number of currently connected computing boards. The video data stream statistics module connected to the computing board is used to perform statistical processing on the video data streams connected to each computing board.
7. The apparatus according to claim 6, characterized in that, The multi-computing board edge computing device further includes: a computing board status display module; The computing board status display module includes a computing board normal status display unit, a computing board idle status display unit, and a computing board not inserted status display unit. The normal status display unit of the computing board is connected to the green indicator light and is used to display the normal status of the computing board through the green indicator light; The computing board idle status display unit is connected to the yellow indicator light and is used to display the idle status of the computing board through the yellow indicator light; The computing board not inserted status display unit is connected to a red indicator light, which is used to display the computing board not inserted status via the red indicator light.