Streaming media management system based on edge intelligent security equipment

Abstract

The invention discloses a streaming media management system based on edge intelligent security and protection equipment, and the system comprises a streaming media management module which is used for achieving the unified access, forwarding and playing, protocol conversion and alarm video synthesis of video streams in the edge intelligent security and protection equipment, and configuring a camera analysis strategy, the streaming media management module supports short video playback of an alarm triggered by a detection algorithm based on a single-frame image, and supports real-time watching and historical playback of a video stream after a detection result of a time sequence type detection algorithm is rendered; the AI analysis docking module communicates with the streaming media management module and an external AI video analysis end through a standardized interface; and the optional configuration item module is used for providing system-level basic configuration and extension configuration. According to the embodiment of the invention, unified access, forwarding and protocol adaptive conversion of multi-source video streams are realized, the concurrent processing and real-time playing performance of multi-channel videos is remarkably improved, automatic identification and storage of alarm events and short video synthesis are realized, and the intelligence and response efficiency of security and protection monitoring are improved.

Description

Technical Field

[0001] This invention relates to the field of computer vision technology, and in particular to a streaming media management system based on edge intelligent security devices. Background Technology

[0002] With the rapid development of internet technology and the increasing demand for video surveillance, streaming media technology has been widely applied in the security field. Traditional video surveillance systems are mostly implemented based on desktop frameworks, characterized by the acquisition, playback, and control of video streams through local clients. However, these systems often suffer from complex deployment, poor cross-platform compatibility, and difficulties in upgrading and maintenance, especially in multi-device, multi-node environments where client installation and version management are extremely costly. Furthermore, as the scale of monitoring continues to expand, traditional desktop architectures struggle to meet the demands of unified management and high-concurrency access in distributed scenarios, severely limiting system flexibility and scalability.

[0003] In recent years, the rapid development of Web technology has brought new solutions to streaming media systems. Web-based streaming media playback and control technologies have gradually matured, enabling real-time transmission, forwarding, and playback of streaming media directly on the browser side. Compared to traditional desktop client methods, Web platforms offer advantages such as no installation required, cross-platform compatibility, ease of updates, and centralized management, significantly reducing system maintenance costs. Furthermore, Web technology inherently supports a B / S architecture, allowing for better interaction with cloud servers, edge nodes, and mobile terminals, making it an important direction for next-generation streaming media management platforms.

[0004] In the field of intelligent security, video surveillance systems not only need to acquire and display video streams, but also need to analyze and understand video content in real time using artificial intelligence algorithms. For example, the introduction of AI algorithms such as facial recognition, target detection, behavior analysis, and abnormal event detection has enabled video surveillance systems to gradually transform from "passive monitoring" to "proactive early warning." However, the traditional cloud-based centralized processing model suffers from problems such as high computational latency, large network bandwidth consumption, and high data privacy risks. Especially in large-scale deployment scenarios, the system is prone to latency and stuttering due to network congestion or excessive cloud load, affecting real-time performance and reliability.

[0005] To address these issues, the introduction of edge computing technology has become a trend. Edge computing, by performing video preprocessing and intelligent analysis on edge devices close to the data source, can effectively reduce the amount of data transmitted upstream, alleviate the computing pressure on cloud servers, and achieve rapid response and localized decision-making. Against this backdrop, streaming media management platforms based on edge devices have emerged. These platforms deploy lightweight streaming media forwarding, caching, and intelligent scheduling modules on edge nodes to achieve local access, forwarding, synthesis, and distribution of video streams, and complete configuration management and remote monitoring through a unified web interface.

[0006] Meanwhile, the concept of microservice architecture has also been widely applied in modern streaming media management platforms. Traditional monolithic application architectures struggle to handle complex scenarios with multiple modules and high concurrency. Microservices, by dividing the system into multiple independent service modules (such as video access, stream forwarding, video analytics, alarm management, user authentication, log monitoring, etc.), enable independent deployment, elastic scaling, and fault tolerance management of these modules. Microservice-based streaming media platforms can achieve inter-module communication through API gateways and asynchronous communication and task scheduling through middleware (such as message queues, Redis caching, Kafka stream processing, etc.), thereby significantly improving the system's concurrency capabilities and overall stability.

[0007] Furthermore, the database layer design is also a crucial component of the streaming media management platform. Security systems contain both structured data (such as device information, user configurations, and alarm logs) and unstructured data (such as video stream clips, screenshots, and model features). Therefore, the system needs to support the collaborative operation of multiple database types. Relational databases (such as MySQL) can be used to store structured data, ensuring data consistency and reliability; while in-memory databases (such as Redis) are used to cache hot data and real-time alarms, improving system response speed. Through a well-designed database architecture, the platform can achieve fast writing and efficient querying of video alarms and event analysis results while ensuring persistence.

[0008] At the streaming media transmission level, modern web streaming media management platforms typically work in conjunction with professional streaming media servers (such as SRS, ZLMediaKit, FFmpeg, etc.) to achieve efficient forwarding and compositing of video streams. Through streaming protocol adaptation (such as RTSP, RTMP, HLS, WebRTC, etc.) and intelligent scheduling mechanisms, the platform can flexibly select the optimal transmission path in different network environments, enabling features such as concurrent playback on multiple terminals, resume interrupted downloads, and low-latency streaming. This flexible streaming media architecture not only meets the high real-time requirements of security systems but also provides a reliable data channel for subsequent intelligent video analysis and data fusion.

[0009] It's worth noting that in actual security deployments, many edge devices (such as front-end cameras, NVRs, and edge computing boxes) lack graphical user interfaces (GUIs) or traditional desktop application environments. Therefore, web-based streaming media management solutions have a natural advantage. Through interactive design and a visual management interface on the web, users can access and configure the system via a browser on any terminal device, while edge nodes only need to run backend services to complete video stream acquisition, forwarding, and intelligent analysis. This "faceless" system architecture enables web streaming media platforms to be efficiently deployed on resource-constrained edge devices, achieving a lightweight, low-power, and remotely manageable operating mode.

[0010] In summary, with the convergence of Web technology, edge computing, and artificial intelligence, traditional desktop-based streaming media management methods are no longer sufficient to meet the real-time, intelligent, and scalable requirements of modern intelligent security systems. Current security monitoring systems face performance bottlenecks and insufficient intelligent collaboration in streaming media management at edge boxes. As the number of monitoring devices increases and video resolution improves, edge nodes encounter problems such as high bandwidth consumption, large transmission latency, and low resource scheduling efficiency during video stream access, forwarding, and display. Furthermore, existing streaming media systems are mostly based on traditional desktop architectures, making it difficult to adapt to edge device environments without graphical interfaces, and hindering efficient integration and data sharing with intelligent video analysis platforms.

[0011] Therefore, it is necessary to improve one or more of the problems existing in the above-mentioned related technical solutions.

[0012] It should be noted that this section is intended to provide background or context for the technical solutions of this disclosure as set forth in the claims. The description herein does not constitute an admission that it is prior art simply because it is included in this section. Summary of the Invention

[0013] The purpose of this invention is to provide a streaming media management system based on edge intelligent security devices, thereby overcoming, at least to some extent, one or more problems caused by the limitations and defects of related technologies.

[0014] This invention first provides a streaming media management system based on edge intelligent security devices, comprising: The streaming media management module, deployed on edge intelligent security devices, is used to realize unified access, forwarding and playback of video streams in edge intelligent security devices. It is also used for protocol conversion, alarm video synthesis and configuration of camera analysis strategies. The alarm video is obtained by the streaming media management module by associating and synthesizing the original video stream and the alarm information from the AI ​​video analysis terminal. The streaming media management module supports short video playback of alarms triggered by detection algorithms based on single-frame images, and also supports real-time viewing and historical playback of video streams rendered after detection results of time-series detection algorithms. The AI ​​analysis interface module communicates with the streaming media management module and the external AI video analysis terminal through a standardized interface. It is used to transmit alarm information and video analysis results generated by the AI ​​video analysis terminal to the streaming media management module, and to transmit the camera analysis strategy configured by the streaming media management module to the AI ​​video analysis terminal, so that the AI ​​video analysis terminal can perform intelligent analysis on the video stream captured by the camera. The optional configuration module provides system-level basic and extended configurations for the streaming media management module and the AI ​​analysis integration module.

[0015] In this invention, the streaming media management module supports converting a single RTSP stream into WebRTC, RTMP, and HLS streaming media protocols.

[0016] In this invention, the streaming media management module writes the basic information of the video stream into a MySQL database for storage. The basic information includes: camera number, original stream address, forwarding stream address, protocol type, and enabled status.

[0017] In this invention, the video analysis results include: alarm type, timestamp, target identifier, and coordinate information.

[0018] In this invention, the streaming media management module persistently stores camera IDs and algorithm mapping information in a MySQL database, and supports simultaneous configuration and priority settings for multiple algorithms.

[0019] In this invention, the streaming media management module includes a front-end and a back-end: the front-end provides a user interface that supports configuring camera information and the original video stream address, selecting to play real-time video / alarm short video, and viewing the synthesized alarm clips; the back-end is deployed in the background service layer of the edge intelligent security device, and is used for unified access, forwarding, and protocol conversion of video streams, associating the original video stream with alarm information, calling tools to synthesize alarm videos, data storage and interface interaction, managing stream status, and synchronizing data to the database.

[0020] The present invention further provides a streaming media management method based on an edge intelligent security device, comprising: S1, the streaming media management module configures the camera analysis strategy, and the AI ​​analysis docking module transmits the camera analysis strategy to the AI ​​video analysis terminal; S2, the AI ​​video analysis end uses the received camera analysis strategy to intelligently analyze the video stream captured by the camera, and generates alarm information and video analysis results; S3, the AI ​​analysis and docking module sends the alarm information and video analysis results to the streaming media management module. The streaming media management module associates and synthesizes the original video stream and alarm information to obtain the alarm video based on the alarm information and video analysis results. S4, the streaming media management module converts the alarm time into a relational data format and stores it in MySQL and the video storage server. It forwards the alarm information to the central server via RPC or MQ through the configured central server address. S5, the streaming media management module plays the alarm video for the user to view.

[0021] In this invention, the streaming media management module includes a front-end and a back-end: the front-end provides a user interface; the back-end calls the built-in streaming media access interface to connect the received video stream to the ZLM streaming media server, realizing real-time forwarding and multi-protocol conversion of the video stream, and the back-end writes the basic information of the video stream into a MySQL database for persistent storage.

[0022] In this invention, the backend is also used to retrieve the WebRTC playback address of the corresponding camera from the MySQL database and return it to the frontend. The frontend calls the WebRTC playback interface through the JavaScript API to realize real-time video playback on the browser side.

[0023] The technical solution provided by this invention may include the following beneficial effects: This invention discloses a streaming media management system based on edge intelligent security devices, which realizes unified access, forwarding, and adaptive protocol conversion of multi-source video streams, significantly improving the concurrent processing and real-time playback performance of multiple video streams. The system can efficiently interface with AI video analysis terminals to achieve automatic identification, storage, and short video synthesis of alarm events, enhancing the intelligence and response efficiency of security monitoring. Through flexible camera algorithm strategy configuration and video management mechanisms, the system can dynamically adjust analysis tasks and storage strategies according to scenario requirements, balancing performance and resource utilization. This invention features a modular and highly scalable overall structure with excellent compatibility and adaptability, and can be widely applied to intelligent security and edge video analysis scenarios. Attached Figure Description

[0024] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0025] Figure 1 This is a schematic diagram of the streaming media management system based on edge intelligent security equipment in this invention; Figure 2 This is a flowchart of the streaming media management method based on edge intelligent security devices in this invention; Figure 3 This is a schematic diagram illustrating the user terminal access to the corresponding camera video stream in this invention; Figure 4 This is a schematic diagram of the backend forwarding video stream via ZLM in this invention; Figure 5 This is a schematic diagram illustrating the millisecond-level delay playback of the video stream in this invention; Figure 6 This is a schematic diagram showing the alarm video clip display results in this invention; Figure 7 This is a schematic diagram illustrating the camera algorithm configuration selection in this invention; Figure 8 This is a schematic diagram of the NTP time calibration configuration in this invention; Figure 9 This is a schematic diagram of the video storage configuration in this invention; Figure 10 This is a schematic diagram of the pedestrian target tracking video in this invention. Detailed Implementation

[0026] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0027] Furthermore, the accompanying drawings are merely illustrative diagrams of embodiments of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities.

[0028] This example implementation first provides a streaming media management system based on edge intelligent security devices. Please refer to [link / reference]. Figure 1 The system includes: The streaming media management module is deployed on edge intelligent security devices to realize unified access, forwarding and playback of video streams in edge intelligent security devices. It is also used for protocol conversion, alarm video synthesis and configuration of camera analysis strategies. The alarm video is obtained by the streaming media management module by associating and synthesizing the original video stream and the alarm information from the AI ​​video analysis terminal. The AI ​​analysis interface module communicates with the streaming media management module and an external AI video analysis terminal through a standardized interface. It is used to transmit alarm information and video analysis results generated by the AI ​​video analysis terminal to the streaming media management module, and to transmit the camera analysis strategy configured by the streaming media management module to the AI ​​video analysis terminal, so that the AI ​​video analysis terminal can intelligently analyze the video stream captured by the camera. The streaming media management module supports short video playback triggered by alarms based on single-frame image detection algorithms, and also supports real-time viewing and historical playback of video streams rendered after detection results of time-series detection algorithms. The optional configuration module provides system-level basic and extended configurations for the streaming media management module and the AI ​​analysis integration module.

[0029] In this embodiment, the present invention constructs a streaming media intelligent management system based on edge intelligent security devices, realizing unified access, forwarding, and adaptive protocol conversion of multi-source video streams, significantly improving the concurrent processing and real-time playback performance of multiple video streams. The system can efficiently interface with AI video analysis terminals to achieve automatic identification, storage, and short video synthesis of alarm events, improving the intelligence and response efficiency of security monitoring. Through flexible camera algorithm strategy configuration and video management mechanisms, the system can dynamically adjust analysis tasks and storage strategies according to scenario requirements, balancing performance and resource utilization. The present invention has a modular overall structure, strong scalability, and good compatibility and adaptability, and can be widely applied to intelligent security and edge video analysis scenarios.

[0030] The specific functions of each module in the above embodiments are explained below.

[0031] 1. Streaming Media Management Module The streaming media management module is used to achieve unified access, forwarding, playback, and protocol conversion management of video streams in edge intelligent security devices. Through unified scheduling and storage management of streaming media data, it achieves efficient streaming media forwarding and distribution, improving the response speed and resource utilization of edge boxes in multi-channel video surveillance scenarios.

[0032] The streaming media management module includes a front-end and a back-end: the front-end provides a user interface that supports configuring camera information and raw video stream addresses, selecting to play real-time video / alarm short videos, and viewing synthesized alarm clips (such as...). Figure 6 (As shown in the figure). The backend is deployed on the edge intelligent security device as a backend service layer, which is used for unified access, forwarding, and protocol conversion of video streams, associating the original video stream with alarm information, calling tools to synthesize alarm videos, data storage and interface interaction, managing stream status, and synchronizing data to the database.

[0033] (1) Adding video streams like Figure 3 As shown, users configure camera information and submit the raw video stream address (such as an RTSP protocol address) to the backend service through the front-end interface. Upon receiving the request, the backend executes the following transaction logic: ①For example Figure 4 As shown, the backend calls the built-in streaming media access interface to connect the RTSP stream to the ZLM streaming media server (ZLMediaKit), realizing real-time forwarding and multi-protocol conversion of the video stream.

[0034] The streaming media management module supports converting a single RTSP stream into multiple streaming media protocols such as WebRTC, RTMP, and HLS to adapt to the playback needs of different terminals and improve the concurrent processing capabilities of the streaming media management system.

[0035] ② The backend writes basic information about the video stream (such as camera ID, original stream address, forwarding stream address, protocol type, and activation status) into a MySQL database for persistent storage. Simultaneously, the streaming media management module supports maintaining the connection status and playback statistics of active streams through Redis caching, ensuring query efficiency is synchronized with the status.

[0036] ③ In the implementation, asynchronous task queues or message middleware (such as RabbitMQ) are used to decouple the stream access events, avoiding the blocking problem when multiple streams are accessed, thereby achieving high concurrency access capability.

[0037] The above solutions enable full automation of the entire process from front-end input to back-end access and multi-protocol forwarding, significantly reducing manual configuration costs; multi-protocol adaptation improves system compatibility, enabling the streaming media management system to simultaneously serve the web, mobile, and AI video analysis ends.

[0038] (2) Video stream playback Once a user selects a camera on the front end, the streaming media management system supports low-latency playback of live video. The specific process is as follows: ① The backend retrieves the WebRTC playback address of the corresponding camera from the MySQL database and returns it to the frontend.

[0039] If the camera's current stream status is disconnected, the backend will automatically trigger the stream restart logic, pulling the video stream again through ZLM to ensure the stability of the video source.

[0040] ② The front-end calls the WebRTC playback interface via the JavaScript API to achieve real-time video playback on the browser. Because WebRTC is based on UDP transmission, streaming media management systems can achieve sub-second (or even millimeter-level) accuracy. Figure 5 (As shown) Low-latency playback meets the timeliness requirements of real-time monitoring.

[0041] ③ The streaming media management system supports controlling video playback permissions through token authentication or session verification mechanisms to prevent unauthorized users from accessing the video stream.

[0042] The above technical solutions enable full-process automation from front-end input to back-end access and multi-protocol forwarding, significantly reducing manual configuration costs; multi-protocol adaptation improves system compatibility, enabling the streaming media management system to simultaneously serve the web, mobile, and AI video analysis ends.

[0043] (3) Video stream synthesis The streaming media management module is used to associate alarm information generated by the AI ​​video analysis terminal with the original video stream, generating short alarm videos that can be viewed quickly (such as...). Figure 10 (As shown). The specific processing flow is as follows: ①The AI ​​video analysis terminal sends alarm information to the streaming media management module via RPC, message queue (MQ) or other middleware, including alarm type, occurrence time and related object identifier.

[0044] ② The streaming media management module retrieves the storage address of the video TS segment corresponding to the alarm event from the MySQL database, and calls the FFmpeg tool to synthesize the relevant segments to generate a short video of the alarm event of about 10 seconds.

[0045] ③ The streaming media management module can support the simultaneous processing of multiple alarms and achieve efficient scheduling of video synthesis through asynchronous task queues, ensuring that short videos can still be generated quickly under multiple concurrent events.

[0046] (4) Camera analysis strategy configuration like Figure 7 As shown, the streaming media management module is used to configure the video stream analysis strategy for each camera, including algorithm selection, detection parameters, time periods, and recording saving strategies: ① Algorithm Selection: Users can specify applicable AI analysis algorithms for each camera, such as object detection, behavior recognition, and abnormal event monitoring. The streaming media management system persistently stores camera IDs and algorithm mapping information in a MySQL database and supports simultaneous configuration of multiple algorithms and priority settings.

[0047] ② Detection parameters and time period: Users can configure algorithm thresholds (such as detection confidence, minimum target size, alarm triggering conditions) and analysis time periods (such as night or specific time windows). The streaming media management system will activate the analysis task at regular intervals according to the configuration to reduce false alarm rate and resource consumption.

[0048] ③ Historical video recording retention strategy: Users can configure the recording retention duration and storage strategy for each camera (edge ​​local, central server, or hybrid storage). The streaming media management system can automatically manage video archiving and rotation according to the strategy, prioritizing the retention of alarm-related video clips.

[0049] Technical effects: The above technical solutions enable refined management of camera analysis tasks, allowing video analysis strategies to be flexibly configured and dynamically adjusted, while optimizing the use of storage and computing resources and improving the efficiency and reliability of intelligent analysis at the edge.

[0050] 2. AI Analysis Integration Module The AI ​​analytics integration module enables efficient data interaction between the streaming media management system and the AI ​​video analytics endpoint, supporting alarm information transmission, analysis result feedback, and intelligent video stream processing. Through this module, edge devices can achieve low-latency, highly reliable data communication with the AI ​​video analytics endpoint, providing support for subsequent video synthesis, alarm generation, and target tracking.

[0051] (1) Alarm information transmission ①The AI ​​video analysis terminal sends the analysis results to the streaming media management module via RPC (Remote Procedure Call), message queue (MQ) or HTTP API, including alarm type, timestamp, target identifier and coordinate information.

[0052] ② After receiving the alarm information, the streaming media management module records the alarm event in the MySQL database and updates the real-time status in the Redis cache to support fast query and alarm triggering.

[0053] ③ The streaming media management system supports asynchronous message processing and batch alarm merging, ensuring efficient alarm processing capabilities in scenarios with multiple targets and multiple cameras. (2) Intelligent video analysis results feedback ① After performing object detection, behavior recognition, or abnormal event analysis on the raw video stream, the AI ​​video analysis terminal sends the results data to the streaming media management module in the form of JSON or binary feature files.

[0054] ② The streaming media management module connects to the AI ​​analysis results through the standardized interface of the AI ​​analysis module and associates them with the video stream to achieve rapid retrieval, video synthesis, and visual annotation of alarm segments.

[0055] ③ For real-time analysis tasks, the streaming media management system can send the analysis results to the front end via WebSocket or push mechanism to achieve real-time display and event annotation.

[0056] The above technical solutions enable intelligent parsing and display of video content, improve the efficiency of streaming media management systems in event identification, tracing, and visualization analysis, and provide reliable data support for subsequent target tracking and behavior analysis.

[0057] 3. Optional Configuration Items Module The optional configuration module is primarily used for basic and extended configuration management of the streaming media management system, providing support for stable access, unified time synchronization, data storage, and remote communication. It enables centralized management and flexible adjustment of system-level parameters through a unified configuration interface or API.

[0058] ① Network Configuration Management: The optional configuration module provides network parameter configuration functions for edge device systems, including setting basic network information such as IP address, subnet mask, gateway, and DNS.

[0059] ②For example Figure 8 As shown, the NTP time calibration function has a built-in NTP (Network Time Protocol) service configuration function, which can perform unified time synchronization through an external NTP server or a local time server to ensure the consistency of timestamps of all devices, video streams and alarm logs, providing an accurate time reference for subsequent event tracing and multi-device linkage.

[0060] ③ For example Figure 9 As shown, the video storage configuration module supports configuring the storage method and strategy for video streams, including local storage path settings, storage capacity threshold management, and recording loop overwrite strategies. The optional configuration module can also configure network storage paths such as FTP / NAS / NFS within the local area network, enabling centralized storage and multi-node access, ensuring the security and traceability of video data.

[0061] ④ Remote Alarm Data Push Configuration: The optional configuration module provides configuration interfaces for various alarm data push methods. Protocols such as HTTP, MQTT, and WebSocket can be selected according to business needs to achieve real-time reporting of alarm information. It supports setting the push frequency, retry mechanism, and authentication information of the target server to ensure reliable alarm message transmission and seamless integration with external systems (such as integrated security platforms).

[0062] ⑤ System Log and Backup Configuration: The optional configuration module provides local and remote storage configuration for the streaming media management system's operating logs, supporting log level filtering and a periodic automatic cleanup mechanism. Users can set the automatic backup cycle and storage location for configuration files and the database, ensuring the streaming media management system's recovery capabilities and data security.

[0063] In summary, the optional configuration module plays a fundamental supporting and operational guarantee role in the entire system architecture, providing a customizable operating environment and a unified system configuration interface for core modules such as intelligent video analysis, alarm processing, and storage management.

[0064] The media management system integrates unified network, time, log, and alarm push configurations, enabling centralized management of system-level parameters and global time synchronization, ensuring data consistency and event traceability under multi-device collaboration.

[0065] This disclosure also provides a streaming media management method based on edge intelligent security devices. Please refer to... Figure 2 The method includes: S1, the streaming media management module configures the camera analysis strategy, and the AI ​​analysis docking module transmits the camera analysis strategy to the AI ​​video analysis terminal; S2, the AI ​​video analysis end uses the received camera analysis strategy to intelligently analyze the video stream captured by the camera, and generates alarm information and video analysis results; S3, the AI ​​analysis and docking module sends the alarm information and video analysis results to the streaming media management module. The streaming media management module associates and synthesizes the original video stream and alarm information to obtain the alarm video based on the alarm information and video analysis results. S4, the streaming media management module converts the alarm time into a relational data format and stores it in MySQL and the video storage server. It forwards the alarm information to the central server via RPC or MQ through the configured central server address. S5, the streaming media management module plays the alarm video for the user to view.

[0066] Similar to the aforementioned system, this method first involves the user configuring the network address corresponding to the camera and adapting it to the appropriate AI algorithm, such as pedestrian intrusion detection or facial recognition, and configuring the corresponding algorithm detection settings. Then, the streaming media management module configures the corresponding algorithm for the camera to the AI ​​intelligent analysis terminal for intelligent camera detection and analysis. Based on the detection JSON message returned by the AI ​​video analysis terminal, the streaming media management module converts the alarm time into a relational data format and stores it in MySQL and the video storage server, mapping the corresponding video path. The corresponding alarm information is forwarded to the central server via RPC and MQ through the configured central server address. When a user views existing alarm video clips, if only the alarm status is being viewed, the video is played directly on the front end. If the user is viewing the tracking results for a specific time zone, the video for that time zone is rendered in real-time using a backup of the detection JSON set and then streamed as an RTSP stream for the user to watch.

[0067] In summary, this application addresses the performance bottlenecks and insufficient intelligent collaboration issues of edge boxes in current security monitoring systems regarding streaming media management, providing an efficient streaming media management method and platform architecture based on edge intelligent security devices. With the increase in the number of monitoring devices and the improvement in video resolution, edge nodes face problems such as high bandwidth consumption, large transmission latency, and low resource scheduling efficiency during video stream access, forwarding, and display. Meanwhile, existing streaming media systems are mostly based on traditional desktop architectures, making it difficult to adapt to edge device environments without graphical interfaces, and also difficult to achieve efficient integration and data sharing with video intelligent analysis platforms. Therefore, this invention proposes a unified and modular streaming media management solution, achieving cross-platform access and remote control through a web-based design, supporting the acquisition, scheduling, display, and caching management of multi-protocol video streams, and enabling localized processing and real-time distribution of video data at the edge. The system also provides a standardized interface to AI intelligent analysis platforms, supporting secondary processing of video streams and task-level interaction, thereby significantly improving the overall performance and flexibility of edge security systems in streaming media processing, intelligent recognition, and event response.

[0068] This application presents a web streaming media management platform based on edge intelligent security devices. Combining microservice architecture, AI video analytics, and multi-database support, it enables efficient management, intelligent analysis, and rapid response of streaming media data in a distributed environment, demonstrating significant application value and development prospects.

[0069] The beneficial effects brought about by the innovation of this invention mainly include: 1. This invention enables multi-protocol forwarding of camera streams by controlling the streaming media server via RPC calls. It supports multiple protocol formats such as WebRTC, FLV, RTSP, and RTMP, achieving millisecond-level latency for user-end monitoring video viewing. Furthermore, it protects user permissions through video token encryption.

[0070] 2. The streaming media management system of this application implements multiple software network interaction methods, such as HTTP, MQ, gRPC, and Redis. Different alarm receiving methods can be constructed according to the traffic, QPS and business needs of the AI ​​analysis terminal. High and low frequency business requests can be handled by switching different interaction methods.

[0071] 3. Customizable device configuration: Users can independently configure the system's NTP service settings, alarm data local storage address (local or local FTP server), historical video storage and overwrite settings, and network interaction methods. The root user can set permissions for other users.

[0072] 4. For time-series algorithms, such as pedestrian tracking algorithms, this system supports real-time viewing of pedestrian tracking trajectories (with global trajectory ID), and also supports reviewing historical pedestrian tracking trajectories.

[0073] It should be noted that although several modules of the system for executing actions are mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more modules described above can be embodied in one module. Conversely, the features and functions of one module described above can be further divided into multiple modules for embodiment. Components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the present invention according to actual needs. Those skilled in the art can understand and implement this without any inventive effort.

[0074] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A streaming media management system based on edge intelligent security devices, characterized in that, include: The streaming media management module, deployed on edge intelligent security devices, is used to realize unified access, forwarding and playback of video streams in edge intelligent security devices. It is also used for protocol conversion, alarm video synthesis and configuration of camera analysis strategies. The alarm video is obtained by the streaming media management module by associating and synthesizing the original video stream and the alarm information from the AI ​​video analysis terminal. The streaming media management module supports short video playback of alarms triggered by detection algorithms based on single-frame images, and also supports real-time viewing and historical playback of video streams rendered after detection results of time-series detection algorithms. The AI ​​analysis interface module communicates with the streaming media management module and the external AI video analysis terminal through a standardized interface. It is used to transmit alarm information and video analysis results generated by the AI ​​video analysis terminal to the streaming media management module, and to transmit the camera analysis strategy configured by the streaming media management module to the AI ​​video analysis terminal, so that the AI ​​video analysis terminal can perform intelligent analysis on the video stream captured by the camera. The optional configuration module provides system-level basic and extended configurations for the streaming media management module and the AI ​​analysis integration module.

2. The streaming media management system based on edge intelligent security equipment according to claim 1, characterized in that, The streaming media management module supports converting a single RTSP stream into WebRTC, RTMP, and HLS streaming media protocols.

3. The streaming media management system based on edge intelligent security equipment according to claim 1, characterized in that, The streaming media management module stores the basic information of the video stream in a MySQL database. This basic information includes: camera number, original stream address, forwarding stream address, protocol type, and enabled status.

4. The streaming media management system based on edge intelligent security equipment according to claim 1, characterized in that, The video analysis results include: alarm type, timestamp, target identifier, and coordinate information.

5. The streaming media management system based on edge intelligent security equipment according to claim 1, characterized in that, The streaming media management module persistently stores camera IDs and algorithm mapping information in a MySQL database, and supports simultaneous configuration and priority settings for multiple algorithms.

6. The streaming media management system based on edge intelligent security equipment according to claim 1, characterized in that, The streaming media management module includes a front-end and a back-end: the front-end provides a user interface that supports configuring camera information and raw video stream addresses, selecting to play real-time video / alarm short video, and viewing synthesized alarm clips; The backend service layer, deployed on edge intelligent security devices, is used for unified access, forwarding, and protocol conversion of video streams, associating original video streams with alarm information, calling tools to synthesize alarm videos, data storage and interface interaction, managing stream status, and synchronizing data to the database.

7. A streaming media management method based on edge intelligent security devices, characterized in that, include: S1, the streaming media management module configures the camera analysis strategy, and the AI ​​analysis docking module transmits the camera analysis strategy to the AI ​​video analysis terminal; S2, the AI ​​video analysis end uses the received camera analysis strategy to intelligently analyze the video stream captured by the camera, and generates alarm information and video analysis results; S3, the AI ​​analysis and docking module sends the alarm information and video analysis results to the streaming media management module. The streaming media management module associates and synthesizes the original video stream and alarm information to obtain the alarm video based on the alarm information and video analysis results. S4, the streaming media management module converts the alarm time into a relational data format and stores it in MySQL and the video storage server. It forwards the alarm information to the central server via RPC or MQ through the configured central server address. S5, the streaming media management module plays the alarm video for the user to view.

8. The streaming media management method based on edge intelligent security equipment according to claim 7, characterized in that, The streaming media management module includes a front-end and a back-end: the front-end provides a user interface; the back-end calls the built-in streaming media access interface to connect the received video stream to the ZLM streaming media server, realizing real-time forwarding and multi-protocol conversion of the video stream, and the back-end writes the basic information of the video stream into a MySQL database for persistent storage.

9. The streaming media management method based on edge intelligent security equipment according to claim 8, characterized in that, The backend is also used to retrieve the WebRTC playback address of the corresponding camera from the MySQL database and return it to the frontend. The frontend calls the WebRTC playback interface through the JavaScript API to realize real-time video playback on the browser.

Images