Video management and control platform integrating multiple video devices

Abstract

The invention relates to a video management control platform integrating multiple video devices. The invention solves the problems of the prior art. The key points of the technical solution are as follows: the platform comprises three layers: a perception layer, a fusion layer and an application layer. A data-oriented architecture is formed from the perception layer from bottom to top. A service-oriented architecture is adopted from top to bottom in the application layer. Data and services are collected and integrated in the fusion layer. The perception layer accesses the video through a video encoding device, an integrated docking national standard video platform and a third-party security alarm system. The fusion layer unifies the video encoding, realizes high-speed transmission, distributed storage, centralized processing and intelligent analysis of video data of the device and video encoding. The application layer is a CS client provided by the platform, or a web client based on the BS architecture of a third-party manufacturer, and a real-time monitoring, history playback, alarm and operation and maintenance application module is constructed based on a standard API interface provided by the video fusion layer.

Description

Technical Field

[0001] The invention relates to a video middleware platform, in particular to a video management and control platform integrating multiple security-related video devices. Background Art

[0002] With the continuous advancement of the digitalization and intelligence of the security industry and the rapid development of IP networks, front-end video technology is maturing, and while costs are falling rapidly, the market demand for applications such as system integration, system networking, and video intelligent analysis is increasing. In the context of video networking, there are many manufacturers and protocols of digital devices on the market. Although organizations such as ONVIF and PSIA advocate an open interface standard to promote the development of IP video equipment, from the current situation, the road to standardization is still a long way to go. At this stage, there are mainly the following problems: First, the front-end manufacturers and equipment interaction protocols are diversified, making it difficult to achieve free access to front-end video; second, the closed nature of the existing video platform cannot effectively cooperate with third-party application systems, and each system runs independently, resulting in information islands; third, due to the particularity of video technology, live broadcast and playback without plug-ins cannot be achieved; fourth, the black screen of the video caused by various abnormalities such as network bandwidth and front-end failure affects the user experience.

[0003] To address the above issues, the video platform is required to have a fully open system architecture that can efficiently and stably integrate various existing video devices, make video images clearer, make the video system more intelligent, cooperate with third-party systems, unify resources, numbering, authority, and management, and ultimately gradually transition from the traditional monitoring model of "post-event evidence collection" to a new management model of "in-process management and pre-event prevention." Summary of the invention

[0004] There are still problems with the above technical solutions. First, the front-end manufacturers and equipment interaction protocols are diverse, making it difficult to achieve free access to the front-end video. Second, the closed nature of the existing video platform makes it impossible to effectively cooperate with third-party application systems. Each system runs independently, and there are information islands. Third, due to the particularity of video technology, live broadcast and playback without plug-ins cannot be achieved. Fourth, due to various abnormalities such as network bandwidth and front-end failures, the black screen of the video affects the user experience. The present invention provides a video management and control platform that integrates multiple security-related video devices.

[0005] The technical solution adopted by the present invention to solve the technical problem is: a video management control platform integrating multiple video devices, including three layers: perception layer, fusion layer and application layer. A data-oriented architecture (DOA) is formed from the perception layer from bottom to top, and a service-oriented architecture (SOA) is adopted from top to bottom in the application layer. Data and services are collected and integrated in the fusion layer.

[0006] The perception layer includes front-end audio and video acquisition equipment, mobile phone / PAD terminals, alarm equipment, and a third-party system platform, and performs video access through video encoding equipment, integrated docking national standard video platform, and third-party security alarm system;

[0007] The fusion layer unifies video coding, realizes high-speed transmission, distributed storage, centralized processing and intelligent analysis of video data of equipment and video coding;

[0008] The application layer is the CS client that comes with the platform, or the web client based on the BS architecture of a third-party manufacturer. The real-time monitoring, historical playback, alarm and operation and maintenance application modules are built based on the standard API interface provided by the video fusion layer to support the rapid configuration and development of third-party systems.

[0009] The present invention builds a middleware service platform based on video monitoring and intelligent analysis, supports distributed, hierarchical deployment architecture and a basic platform oriented to SOA service architecture, integrates various video encoding devices, intelligent analysis devices, and national standard platforms, shields the differences between front-end devices and platforms, provides a standardized interface based on HTTP protocol and a solution for playing videos without plug-ins, and provides complete video business support for third-party systems.

[0010] Preferably, the fusion layer includes a service layer, a data layer and a transport layer. The transport layer is used to receive data from the perception layer and to carry network resources for transmitting video and alarm data.

[0011] The data layer receives the data from the transport layer for processing, and provides a unified video and alarm data access portal for all services, targeting video unstructured data and alarm structured data;

[0012] The service layer receives data from the data layer for processing and use, and provides basic services and interface services.

[0013] Preferably, the basic services include streaming media services, video storage services, distributed management services, cascade management services, resource scheduling services, device control services and streaming media live broadcast services;

[0014] The interface service is based on a basic service architecture and is a software service with unified specifications.

[0015] Preferably, the video management and control platform integrating various video devices includes a load balancing server, a central management server, a device management server, a streaming media server, a database and a live broadcast server.

[0016] As a preferred option, device access adopts a modular design and accesses the device SDKs of mainstream manufacturers such as Hikvision and Dahua. It also supports ONVIF, RTSP, and GB28181 standard protocol device access to ensure the stability of basic services. Devices with other private protocols access the platform through a proxy framework.

[0017] As a preferred option, a complete secondary development interface based on the HTTP protocol is provided. The protocol content involves user login, users, equipment, organizational structure management, real-time video, video playback, PTZ control, alarm notification, query and alarm linkage, and realizes data interoperability and information sharing with the third system.

[0018] As a preferred option, the video middleware provides plug-in-free stream playback based on H5 technology. The third-party system obtains the video playback path address from the platform and uses a QR code to share it to WeChat and Alipay applets. The plug-in-free stream playback is compatible with all mainstream terminals on the market, including Windows PC, MAC, Android and IOS terminals.

[0019] As a preferred option, the device management service and media forwarding service of the video middleware platform are both horizontally expanded, and dynamic load balancing is achieved according to the server configuration. The service background dynamically allocates camera forwarding in real time according to the service operation resources to achieve smooth expansion of the service.

[0020] As a preferred option, through the cascade deployment architecture, users can actively add subordinate platforms and register subordinate platforms without public fixed IPs, deploy a 5-level platform deployment, accept alarms from third-party systems, and link camera recording, pan / tilt control, and device control outputs within the platform according to the alarm definition, and notify the platform's CS client.

[0021] Preferably, a detection algorithm is used to determine in advance whether there is an abnormality in the camera. If 80% of the abnormalities occur, the user is prompted with the cause of the abnormality in advance, and the black screen is reduced or skipped directly according to the user's selection habits.

[0022] Preferably, after the detection algorithm determines in advance whether the camera is abnormal, the following steps are performed:

[0023] S1, mark the start time S and end time E of the black screen, and then determine the length of the black screen time. If the length of the black screen time is less than the set value, then execute step S2 according to the user's choice. If the length of the black screen time is greater than or equal to the set value, then directly indicate the alarm content;

[0024] S2, if the user chooses the automatic supplement processing mode, then execute step S3; if the user chooses the automatic switching processing mode, then directly switch to the picture of the time frame E and freeze it, wait for the playback progress to reach time E naturally, and then jump to execute step S4;

[0025] S3, playing the images between the start time S and the end time E by inserting frames until the playback progress reaches the time E naturally;

[0026] S4, play the subsequent pictures normally;

[0027] In the step S3, the content of inserting the frame is to search for n frames before the start time S, select the frame closest to the E+1 frame in the n frames and mark it as the e frame, select the frame closest to the S-1 frame in the n frames and mark it as the s frame, if there is no frame closest to the set frame in the n frames, directly switch to the time E frame and freeze it, wait for the playback progress to reach the time E naturally, and then jump to execute step S4; if there are s frames and e frames, execute the following steps, if the playback order between the s frames and the e frames is in positive order, directly add the s frames to the e frames to the black screen between the start time S and the end time E, if the playback order between the s frames and the e frames is in reverse order, adjust the playback order of the s frames to the e frames to be in reverse order and add them to the black screen. Between the start time S and the end time E; according to the duration of the start time S and the end time E, the s-frame picture is adapted to the e-frame picture, and the adaptation method is: if the fixed number of frames according to the duration is less than the number of frames from the s-frame picture to the e-frame picture, the number of frames required to be extracted from the s-frame picture to the e-frame picture is calculated, and then the frames are extracted from the s-frame picture to the e-frame picture at uniform intervals; if the fixed number of frames according to the duration is greater than the number of frames from the s-frame picture to the e-frame picture, the time required to be filled in the number of frames from the s-frame picture to the e-frame picture is calculated, and then the frame reduction playback is performed in proportion to the s-frame picture to the e-frame picture; if the number of frames after the frame reduction playback is less than the set number of frames per second, the freeze frame is inserted from the s-frame picture to the e-frame picture at uniform intervals, and the duration of the freeze frame is less than the set value.

[0028] Preferably, in the step of selecting the e-frame or the s-frame, the method of selecting the frame with the closest picture includes the following sub-steps: Z1, determining a template picture, the template picture being a frame with the largest number of occurrences of which the overlap degree is greater than a set value in all recorded videos;

[0029] Z2, in the E+1 frame or S-1 frame picture, select the difference picture from the template picture for comparison, select several areas at the edge of the difference picture as several target comparison areas to be compared, the edge of the difference picture refers to a picture of a set shape with the inside filled with the difference picture and the outside filled with a pure color;

[0030] Z3, calculate the number of times the target contrast area appears in the n frames for selection and the displacement value when it appears, set a positive weighted value for the number of times the target contrast area appears, set another negative weighted value for the displacement value when it appears, and calculate the weighted total according to the set weighted values. Only when the positive weighted value, the negative weighted value and the comprehensive weighted value all meet the set frame is the closest frame, otherwise it is determined that there is no e frame or s frame.

[0031] The substantial effect of the present invention is: the present invention constructs a middleware service platform based on video monitoring and intelligent analysis, supports a distributed, hierarchical deployment architecture and a basic platform oriented to the SOA service architecture, integrates various video encoding devices, intelligent analysis devices, and national standard platforms, shields the differences between front-end devices and platforms, provides a standardized interface based on the HTTP protocol and a solution for playing videos without plug-ins, and provides complete video business support for third-party systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0033] Figure 2 It is a schematic diagram of the topological structure of part of the present invention; DETAILED DESCRIPTION

[0034] The technical solution of the present invention is further described in detail below through specific embodiments.

[0035] Embodiment 1:

[0036] A video management and control platform that integrates multiple video devices, including the perception layer, fusion layer, and application layer. From the perception layer, a data-oriented architecture (DOA) is formed from bottom to top, and the application layer adopts a service-oriented architecture (SOA) from top to bottom. Data and services are integrated in the fusion layer.

[0037] The perception layer includes front-end audio and video acquisition equipment, mobile phone / PAD terminals, alarm equipment, and a third-party system platform, and performs video access through video encoding equipment, integrated docking national standard video platform, and third-party security alarm system;

[0038] The fusion layer unifies video coding, realizes high-speed transmission, distributed storage, centralized processing and intelligent analysis of video data of equipment and video coding;

[0039] The application layer is the CS client that comes with the platform, or the web client based on the BS architecture of a third-party manufacturer. The real-time monitoring, historical playback, alarm and operation and maintenance application modules are built based on the standard API interface provided by the video fusion layer to support the rapid configuration and development of third-party systems.

[0040] This embodiment builds a middleware service platform for video monitoring and intelligent analysis, supports distributed, hierarchical deployment architecture and a basic platform oriented to SOA service architecture, integrates various video encoding devices, intelligent analysis equipment, and national standard platforms, and shields the differences between front-end devices and platforms. It provides a standardized interface based on the HTTP protocol and a solution for playing videos without plug-ins, providing complete video business support for third-party systems. The fusion layer includes a service layer, a data layer, and a transport layer. The transport layer is used to receive data from the perception layer and carry network resources for video and alarm data transmission.

[0041] The data layer receives the data from the transport layer for processing, and provides a unified video and alarm data access portal for all services, targeting video unstructured data and alarm structured data;

[0042] The service layer receives data from the data layer for processing and use, and provides basic services and interface services.

[0043] The basic services include streaming media services, video storage services, distributed management services, cascade management services, resource scheduling services, device control services and streaming media live broadcast services;

[0044] The interface service is based on a basic service architecture and is a software service with unified specifications.

[0045] The video management and control platform integrating various video devices includes a load balancing server, a central management server, a device management server, a streaming media server, a database and a live broadcast server.

[0046] In this embodiment, the device access adopts a modular design, accesses the device SDKs of mainstream manufacturers Hikvision and Dahua, and supports ONVIF, RTSP, and GB28181 standard protocol device access to ensure the stability of basic services; other private protocol devices access the platform through the proxy framework.

[0047] This embodiment provides a complete secondary development interface based on the HTTP protocol. The protocol content involves user login, users, devices, organizational structure management, real-time video, video playback, PTZ control, alarm notification, query and alarm linkage, and realizes data intercommunication and information sharing with the third system.

[0048] The video middleware of this embodiment provides plug-in-free code stream playback based on H5 technology. The third-party system obtains the video playback path address from the platform and uses the QR code to share it to WeChat and Alipay applets. The plug-in-free code stream playback is compatible with all mainstream terminals on the market, including Windows PC, MAC, Android and IOS terminals.

[0049] The video middleware platform device management service and media forwarding service of this embodiment are both horizontally expanded, and dynamic load balancing is achieved according to the configuration of the server. The service background dynamically allocates camera forwarding according to the service operation resources in real time to achieve smooth expansion of the service. This embodiment uses a cascaded deployment architecture, and users can actively add subordinate platforms and register subordinate platforms without public network fixed IPs. There are 5-level platform deployments, accepting alarms from third-party systems, and can link camera recording, pan-tilt control and device control outputs in the platform according to the alarm definition, and notify the CS client of the platform. This embodiment uses a detection algorithm to pre-determine whether the camera is abnormal. If 80% of the abnormalities occur, the user is prompted in advance for the abnormal reason. According to the user's selection habits, the black screen is reduced or the black screen is directly skipped. The detection algorithm in this embodiment can be a variety of existing detection algorithms. The simplest algorithm is to detect the number of all-black or all-white images in subsequent frames. If the number of all-black or all-white images is greater than 80%, it is an abnormal frame.

[0050] After the detection algorithm determines in advance whether the camera is abnormal, perform the following steps:

[0051] S1, mark the start time S and end time E of the black screen, and then determine the length of the black screen time. If the length of the black screen time is less than the set value, then execute step S2 according to the user's choice. If the length of the black screen time is greater than or equal to the set value, then directly indicate the alarm content;

[0052] S2, if the user chooses the automatic supplement processing mode, then execute step S3; if the user chooses the automatic switching processing mode, then directly switch to the picture of the time frame E and freeze it, wait for the playback progress to reach time E naturally, and then jump to execute step S4;

[0053] S3, playing the images between the start time S and the end time E by inserting frames until the playback progress reaches the time E naturally;

[0054] S4, play the subsequent pictures normally;

[0055] In the step S3, the content of inserting the frame is to search for n frames before the start time S, select the frame closest to the E+1 frame in the n frames and mark it as the e frame, select the frame closest to the S-1 frame in the n frames and mark it as the s frame, if there is no frame closest to the set frame in the n frames, directly switch to the time E frame and freeze it, wait for the playback progress to reach the time E naturally, and then jump to execute step S4; if there are s frames and e frames, execute the following steps, if the playback order between the s frames and the e frames is in positive order, directly add the s frames to the e frames to the black screen between the start time S and the end time E, if the playback order between the s frames and the e frames is in reverse order, adjust the playback order of the s frames to the e frames to be in reverse order and add them to the black screen. Between the start time S and the end time E; according to the duration of the start time S and the end time E, the s-frame picture is adapted to the e-frame picture, and the adaptation method is: if the fixed number of frames according to the duration is less than the number of frames from the s-frame picture to the e-frame picture, the number of frames required to be extracted from the s-frame picture to the e-frame picture is calculated, and then the frames are extracted from the s-frame picture to the e-frame picture at uniform intervals; if the fixed number of frames according to the duration is greater than the number of frames from the s-frame picture to the e-frame picture, the time required to be filled in the number of frames from the s-frame picture to the e-frame picture is calculated, and then the frame reduction playback is performed in proportion to the s-frame picture to the e-frame picture; if the number of frames after the frame reduction playback is less than the set number of frames per second, the freeze frame is inserted from the s-frame picture to the e-frame picture at uniform intervals, and the duration of the freeze frame is less than the set value.

[0056] In the step of selecting the e frame or the s frame, the method for selecting the frame with the closest picture includes the following sub-steps: Z1, determining a template picture, where the template picture is a frame with the largest number of occurrences in all recorded videos and whose overlap degree is greater than a set value; Z2, in the E+1 frame or the S-1 frame, comparing the template picture to select a difference picture therein, and selecting a plurality of areas at the edge of the difference picture as a plurality of target comparison areas to be compared, where the edge of the difference picture refers to a picture of a set shape whose inner side is filled with the difference picture and whose outer side is filled with a pure color;

[0057] Z3, calculate the number of times the target contrast area appears in the n frames for selection and the displacement value when it appears, set a positive weighted value for the number of times the target contrast area appears, set another negative weighted value for the displacement value when it appears, and calculate the weighted total according to the set weighted values. Only when the positive weighted value, the negative weighted value and the comprehensive weighted value all meet the set frame is the closest frame, otherwise it is determined that there is no e frame or s frame. For example, in this embodiment, 5 edge images are selected as the target comparison area, then all the difference pictures in the n frames are checked to see whether there are 5 edge images in the difference pictures. If there is no edge image, the weight value is 0.18, and if there are at least 4 edge images, it is qualified. Then the displacement value of each edge image is calculated, and the weight of each unit of the displacement value is -0.02. All displacement values ​​that are greater than the set value after weighting are qualified. When all pictures that meet the above conditions are summed up, the positive and negative weighted values ​​are the comprehensive weighted value. A frame corresponding to the maximum comprehensive weighted value is selected as the closest frame to the target, and then subsequent operations are performed. Since the existence of the s frame is easy to obtain, in general, the S-1 frame before the S frame can also be directly used as the s frame, which is generally no problem. The e frame is more complicated, and generally there may be no e frame, or the acquisition of the e frame is difficult. Therefore, for the e frame, the weighted value can be further adjusted or the selected threshold value can be adjusted. In short, the algorithms for selecting the s frame and the e frame can be the same, but the weighted value can be adjusted. The interpolation method in this embodiment is mainly suitable for repeated movements in a short period of time. For example, if a person is wandering under the camera, there is no need to play it in black screen mode when a camera failure occurs. It is only necessary to repeat the previous wandering picture. Furthermore, there are no other objects under the camera. The black screen such as the short-term fallen leaves blocking the camera can also be covered by this method, thereby improving the user experience. However, a long black screen may trigger a failure, so a warning message is required. Of course, since the camera is evidential, it is also necessary to mark the S frame and the E frame when used as evidence. Therefore, in the process of interpolation, a reminder of non-real-time pictures can be added.

[0058] This system adopts comprehensive security protection measures and has a high degree of security and confidentiality. Strict security isolation and protection are carried out on external data accessing the system to ensure the security of access. The system supports backup and redundancy measures for key data and key program modules, has strong fault tolerance and system recovery capabilities, and ensures the long-term normal operation of the system. This system adopts telecommunications-level reliability design. With the support of dual-machine hot standby, automatic load balancing, storage control functions and the adaptability of multiple networks, the platform runs stably and has a low failure rate. This system can be deployed in a private or hybrid cloud, supports cluster deployment, and effectively tolerates disasters and faults to ensure system performance and stability. Multiple heavy-load network elements such as device access, user access, and video distribution on the platform can be deployed at the same time. They work in coordination under the scheduling of core scheduling and load balancing services. Any problem will not affect the stability of the entire system. The system supports local hot standby and remote backup. Once a downtime occurs, the system can be restored from the breakpoint in time. The system is based on modular, distributed, and hierarchical design concepts. It can be developed in stages. The coupling between modules is low, which is conducive to adjustment, upgrading, and secondary development. It supports the design and development of distributed transactions, supports automated deployment, and has good scalability. This system adopts the mainstream architecture, with back-end services and front-end UI separated. The video middleware provides service requests based on the RESTFUL HTTP interface and supports various terminals such as PCs and mobile phones. It provides video-related business support to other heterogeneous service systems through standard Internet services. With the support of distributed design, cascade processing, step-by-step distribution strategy, and platform intelligent management, the system supports the access of millions of devices and users. The expansion of the system scale can be implemented by adding servers, service units, and sub-platforms. The platform adopts a distributed and self-balancing architecture. Under the condition of ensuring network interconnection, each service in the system can be installed and run independently; adding services will not affect the operation of the system, and the original services can run as usual. The system has independent integrated basic management and can be connected with other systems, including but not limited to security platform systems. It can support secondary development of the system. Each system adopts a standard data interface and has the ability to exchange and share data with other information systems.

[0059] The above-described embodiment is only a preferred solution of the present invention and does not limit the present invention in any form. There are other variations and modifications without exceeding the technical solution described in the claims.

Claims

1. A video management and control platform integrating multiple video devices, characterized in that: It includes three layers: perception layer, fusion layer, and application layer. From the perception layer, a data-oriented architecture (DOA) is formed from bottom to top, and the application layer adopts a service-oriented architecture (SOA) from top to bottom. Data and services are integrated in the fusion layer. The perception layer includes front-end audio and video acquisition equipment, mobile phone / PAD terminals, alarm equipment, and a third-party system platform, and performs video access through video encoding equipment, integrated docking national standard video platform, and third-party security alarm system; The fusion layer unifies video coding, realizes high-speed transmission, distributed storage, centralized processing and intelligent analysis of video data of equipment and video coding; The application layer is the CS client that comes with the platform, or the web client based on the BS architecture of a third-party manufacturer. The real-time monitoring, historical playback, alarm and operation and maintenance application modules are built based on the standard API interface provided by the video fusion layer to support the rapid configuration and development of third-party systems; The fusion layer includes a service layer, a data layer and a transport layer. The transport layer is used to receive data from the perception layer and to carry network resources for video and alarm data transmission. The data layer receives the data from the transport layer for processing, and provides a unified video and alarm data access portal for all services, targeting video unstructured data and alarm structured data; The service layer receives data from the data layer for processing and use, and provides basic services and interface services; The basic services include streaming media services, video storage services, distributed management services, cascade management services, resource scheduling services, device control services and streaming media live broadcast services; The interface service is a software service with unified specifications based on the basic service architecture; Use the detection algorithm to determine in advance whether the camera has any abnormality. If 80% of the abnormality occurs, the user will be prompted with the cause of the abnormality in advance. According to the user's choice habits, the black screen will be reduced or directly skipped; After the detection algorithm determines in advance whether the camera is abnormal, perform the following steps: S1, mark the start time S and end time E of the black screen, and then determine the length of the black screen time. If the length of the black screen time is less than the set value, then according to the user's choice, execute step S2. If the length of the black screen time is greater than or equal to the set value, then directly indicate the alarm content; S2, if the user chooses the automatic supplement processing mode, then execute step S3; if the user chooses the automatic switching processing mode, then directly switch to the picture of the time frame E and freeze it, wait for the playback progress to reach time E naturally, and then jump to execute step S4; S3, playing the images between the start time S and the end time E by inserting frames until the playback progress reaches the time E naturally; S4, play the subsequent pictures normally; In the step S3, the content of inserting the frame is to search for n frames before the start time S, select the frame closest to the E+1 frame in the n frames and mark it as the e frame, select the frame closest to the S-1 frame in the n frames and mark it as the s frame, if there is no frame closest to the set frame in the n frames, directly switch to the time E frame and freeze it, wait for the playback progress to reach the time E naturally, and then jump to execute step S4; if there are s frames and e frames, execute the following steps, if the playback order between the s frames and the e frames is in positive order, directly add the s frames to the e frames to the black screen between the start time S and the end time E, if the playback order between the s frames and the e frames is in reverse order, adjust the playback order of the s frames to the e frames to be in reverse order and add them to the black screen. Between the start time S and the end time E; according to the duration of the start time S and the end time E, the s-frame picture is adapted to the e-frame picture, and the adaptation method is: if the fixed number of frames according to the duration is less than the number of frames from the s-frame picture to the e-frame picture, the number of frames required to be extracted from the s-frame picture to the e-frame picture is calculated, and then the frames are extracted from the s-frame picture to the e-frame picture at uniform intervals; if the fixed number of frames according to the duration is greater than the number of frames from the s-frame picture to the e-frame picture, the time required to be filled in the number of frames from the s-frame picture to the e-frame picture is calculated, and then the frame reduction playback is performed in proportion to the s-frame picture to the e-frame picture; if the number of frames after the frame reduction playback is less than the set number of frames per second, the freeze frame is inserted from the s-frame picture to the e-frame picture at uniform intervals, and the duration of the freeze frame is less than the set value.

2. The video management control platform integrating multiple video devices according to claim 1, characterized in that: The video management and control platform integrating various video devices includes a load balancing server, a central management server, a device management server, a streaming media server, a database and a live broadcast server.

3. The video management control platform integrating multiple video devices according to claim 1, characterized in that: Device access adopts a modular design and access device SDK, and supports ONVIF, RTSP, and GB28181 standard protocol device access to ensure the stability of basic services; other private protocol devices access the platform through the proxy framework.

4. The video management control platform integrating multiple video devices according to claim 1, characterized in that: Provides a complete secondary development interface based on HTTP protocol. The protocol content involves user login, users, equipment, organizational structure management, real-time video, video playback, PTZ control, alarm notification, query and alarm linkage, and realizes data intercommunication and information sharing with the third system.

5. The video management control platform integrating multiple video devices according to claim 1, characterized in that: The video middleware provides plug-in-free stream playback based on H5 technology. The third-party system obtains the video playback path address from the platform and uses the QR code to share it to WeChat and Alipay applets. The plug-in-free stream playback is compatible with all mainstream terminals on the market, including Windows PC, MAC, Android and IOS terminals.

6. The video management control platform integrating multiple video devices according to claim 1, characterized in that: In the step of selecting the e-frame or the s-frame, the method of selecting the frame with the closest picture includes the following sub-steps: Z1, determine the template picture, which is the frame with the largest number of occurrences of overlap greater than the set value in all recorded videos; Z2, in the E+1 frame or S-1 frame picture, select the difference picture from the template picture for comparison, select several areas at the edge of the difference picture as several target comparison areas to be compared, the edge of the difference picture refers to a picture of a set shape with the inside filled with the difference picture and the outside filled with a pure color; Z3, calculate the number of times the target contrast area appears in the n frames for selection and the displacement value when it appears, set a positive weighted value for the number of times the target contrast area appears, set another negative weighted value for the displacement value when it appears, and calculate the weighted total according to the set weighted values. Only when the positive weighted value, the negative weighted value and the comprehensive weighted value all meet the set frame is the closest frame, otherwise it is determined that there is no e frame or s frame.

Images