WO2013117065A1 - Video processing method and system, ivw and ivu - Google Patents

Abstract

Disclosed are a video processing method and system, intelligent video management platform (IVM) and intelligent video processing unit (IVU), the system comprising the IVM and the IVU; the IVW is used to store and manage the capability set of administered IVUs, and according to the request of a video monitoring system, and from the administered IVUs, to select and invoke an IVU satisfying the capability requirement; the IVU is used to accept the management and scheduling of the IVW, to analyze and recognize the image or video stream inputted into the IVU from the video monitoring system, and to output the processed result to the IVW. The present invention superimposes a video analysis and recognition technology on the basis of an existing video monitoring system, thus providing video analysis and recognition service in a flexible and configurable manner.

Description

 Video processing method and system, IVW and IVU technical field

 The present invention relates to video surveillance technology and intelligent video analysis technology in the information field, and in particular to a video processing method and system, IVW and IVU. Background technique

 At present, the video surveillance system is an important service for operators. The video surveillance system is deployed in large quantities as the basis for services such as urban security, security protection, and remote monitoring.

 Intelligent Video (IV) technology is derived from the research of Computer Vision (CV) and Artificial Intelligence (AI). Its development goal is to establish a mapping relationship between images and event descriptions. Identify and identify key target objects from a wide range of video images. This research applied to the security video surveillance system will be able to use the powerful data processing capabilities of the computer to filter out useless or interference information in the image, automatically analyze and extract key useful information in the video source, so that the camera in the traditional monitoring system not only Being a human eye also makes the "smart video analysis" computer a human brain and has a more "smart" way of learning and thinking.

 With the further enhancement of video analysis and processing technology, the existing related technologies have been able to accurately identify and analyze video content. However, how to superimpose the video analysis and recognition technology based on the existing video surveillance system to provide the video analysis and recognition service in a flexible and configurable manner is an urgent problem to be solved. Summary of the invention

In view of this, the main object of the present invention is to provide a video processing method and system, IVW and IVU, to implement superimposed video analysis and recognition technology based on an existing video monitoring system. Video analytics identification services are now available in a flexible, configurable manner.

 In order to achieve the above object, the technical solution of the present invention is achieved as follows:

 The invention provides a video processing system, comprising: an intelligent management platform IVM and an intelligent video processing unit IVU,

 The IVW is used to save and manage the capability set of the IVU under the jurisdiction, and according to the request of the video monitoring system, select an IVU that meets the capability requirement from the IVU under the jurisdiction and call it;

 The IVU is configured to accept management and scheduling of the IVW, analyze and identify an image or video stream input from the video surveillance system, and output a processing result to the IVW.

 The IVW is further configured to provide an intelligent application policy configuration service, and save the configured intelligent application policy.

 The IVW is further configured to receive a call from the video surveillance system to the IVW-supplied smart application policy, and feed back the address of the IVU of the execution policy to the video surveillance system according to the saved IVU capability set.

 The IVU is further configured to: when an IVU as an execution policy is invoked by the IVW, acquire an image or a video stream from a destination indicated by the video monitoring system, analyze and identify an input image or a video stream, and The processing result is output to the IVW.

 The invention also provides an intelligent management platform IVW, comprising:

 The IVU management scheduling module is configured to save and manage the capability set of the intelligent video processing unit IVU, and select an IVU that meets the capability requirement from the IVU under the control according to the request of the video monitoring system;

 The analysis result saving module is configured to receive an output result of the IVU analyzing and identifying the image or video stream input to the IVU from the video monitoring system, and save the result.

The IVW further includes: a policy management module, configured to provide a smart application policy configuration service, and save the configured smart application policy. The policy management module is further configured to: accept a call of the video monitoring system to the IVW saved smart application policy, and notify the IVU management scheduling module;

 The IVU management scheduling module is further configured to: feed back an address of the IVU of the execution policy to the video monitoring system according to the invoked smart application policy and the saved capability set of the IVU.

 The invention also provides an intelligent video processing unit IVU, comprising:

 The capability reporting module is configured to report the capability set of the IVU to the IVW, and accept management and scheduling of the IVW;

 And a processing module, configured to analyze and identify an image or video stream input to the IVU from the video monitoring system, and output the processing result to the IVW.

 The processing module is further configured to: when the IVU is invoked by the IVW as an execution policy, acquire an image or a video stream from a destination indicated by the video monitoring system, analyze the input image or the video stream, and The processing is identified, and the processing result is output to the IVW.

 The invention also provides a video processing method, the method comprising:

 The intelligent management platform IVW saves and manages the capability set of the intelligent video processing unit IVU under its jurisdiction, and selects and calls the IVU that satisfies the capability requirements from the IVUs under the jurisdiction of the video surveillance system;

The IVU accepts the management and scheduling of the IVW, analyzes and identifies the image or video stream input to the IVU from the video surveillance system, and outputs the processing result to the IVW.

 The method further includes: the IVW providing an intelligent application policy configuration service, and saving the configured smart application policy.

 The method further includes: the IVW accepting a call from the video surveillance system to the IVW-preserved intelligent application policy, and feeding back the address of the IVU of the execution policy to the video surveillance system according to the saved IVU capability set.

The method further includes: the IVU executing the policy acquiring an image or video stream from the destination indicated by the video surveillance system, analyzing and identifying the input image or video stream, and outputting the processing result to the IVW. The intelligent application policy saved by the IVW is called by the video surveillance platform CMS inside the video surveillance system, and the IVW feeds back the address of the IVU of the execution policy to the CMS, and the destination indicated by the CMS is the front-end PU inside the video surveillance system;

 Correspondingly, the IVU obtains an image or video stream from the PU indicated by the CMS as:

 The IVU requests an image or video stream of the PU from a CMS, and the CMS forwards the request to a corresponding PU;

 The PU returns a response message to the CMS, the CMS selects a video distribution unit VTDU that forwards an image or video stream, and sends the response message to the IVU;

 The CMS notifies the selected VTDU of the address of the PU transmitting the image or video stream and the address of the IVU receiving the image or video stream;

 A session is established between the PU and the IVU through the VTDU, and the image or video stream of the PU is forwarded to the IVU via the VTDU.

 The intelligent application policy saved by the IVW is called by the CMS inside the video surveillance system, and the address of the IVU of the IVW feedback execution policy is given to the CMS, and the destination indicated by the CMS is the PU inside the video monitoring system;

 Correspondingly, the IVU obtains an image or video stream from the PU indicated by the CMS as:

 The IVU requests an image or video stream of the PU from a CMS, and the CMS forwards the request to a corresponding PU;

 The PU returns a response message to the CMS, and the CMS sends the response message to the IVU;

A session is established between the PU and the IVU, and the IVU receives an image or video stream sent by the PU. The video processing method and system, the IVW and the IVU provided by the invention provide a video analysis and identification service in a flexible and configurable manner on the basis of the existing video monitoring system; and can flexibly select different video front end PUs The video content is analyzed and identified, and the video analysis and recognition capabilities and results are uniformly provided in a flexible and configurable manner based on the existing video surveillance system. Video processing service. DRAWINGS

 1 is a schematic structural diagram of a video processing system according to an embodiment of the present invention;

 2 is a flowchart of registering an IVU with an IVM in an embodiment of the present invention;

 3 is a flowchart of a smart application policy configuration according to an embodiment of the present invention;

 4 is a flowchart of a video processing method according to an embodiment of the present invention;

 FIG. 5 is a flowchart of acquiring an video surveillance image by an IVU according to an embodiment of the present invention; FIG.

 6 is a schematic structural diagram of an IVW according to an embodiment of the present invention;

 FIG. 7 is a schematic structural diagram of an IVU according to an embodiment of the present invention. detailed description

 The technical solutions of the present invention are further elaborated below in conjunction with the accompanying drawings and specific embodiments. A video processing system according to an embodiment of the present invention, as shown in FIG. 1, mainly includes: an intelligent management platform (IVM) and an intelligent video processing unit (IVU). Among them, IVW is mainly used to save and manage the capability set of the IVU under its jurisdiction. According to the request of the video surveillance system, the IVU that meets the capability requirements is selected and called from the IVUs under its jurisdiction. The IVU is mainly used to accept the management and scheduling of the IVW, analyze and identify the image or video stream input from the video surveillance system, and output the processing result to the IVW.

 Preferably, the IVW is further configured to provide an intelligent application policy configuration service, and save the configured smart application policy. Accept the IVU capability query and/or intelligent application policy query request of the video surveillance system, and feedback the query result to the video surveillance system.

The IVW accepts the call of the video monitoring system to the intelligent application policy saved by the IVW, and feeds back the address of the IVU of the execution strategy to the video monitoring system according to the saved IVU capability set. When the IVU as the execution strategy is called by the IVW, the image or video stream is acquired from the destination indicated by the video surveillance system, and the input image or video stream is analyzed and identified, and the processing knot is output. Give IVW. The results of the image or video stream analysis and recognition process can also be output to the video surveillance system.

 The video surveillance system mainly includes: a video surveillance platform (CMS), a video distribution unit (VTDU), a front end (PU), a client (CU), and a network storage unit (NRU).

 Among them, CMS is the central management server of video surveillance system, which implements functions such as client/user management, front-end/platform equipment management, call control, management and service parameter configuration, content distribution, billing, virtual domain management and business process control. It can interface with 97/CRM system and billing system.

 The VTDU is connected to the CMS, which is a media forwarding/distribution unit of the video surveillance system. It implements functions such as audio/video reception, forwarding, distribution, and billing data collection, and optionally implements multicast functions. Multiple levels of cascading, distributed deployment.

 NRU connection VTDU is a network video storage unit in the video surveillance system, which implements functions such as receiving/storing audio/video, recording storage management, storage medium management, and video playback service.

The NRU must be able to connect to various physical storage devices such as Storage Area Network (SAN), Network Attached Storage (NAS), and Direct Attached Storage (DAS). NRU enables distributed deployment. NRU accepts the management of CMS.

 The PU connects the CMS and the VTDU. It is the information collection terminal of the video surveillance system. It implements the collection function of video information, audio information, data information and alarm information, and has the bidirectional transmission function of voice information and data information. Some PUs can realize the storage function of audio and video recording. PUs usually include the following devices: front-end devices (video servers or DVRs, etc.), cameras, pan/tilt devices, alarm input and output devices, and so on.

The CU connects to the VTDU, which is the client application end of the video surveillance system, and realizes the presentation of video information, audio information, data information and alarm information to the user. Depending on the type of client, some CUs also have the functions of device management and user management for the video surveillance system. In the embodiment of the present invention, the CU can implement the configuration of the IVM, and can obtain the result after the IVU processing. The functions of the IVM and IVU are further described below.

 The IVM is connected to the IVU and CMS, which is a central service platform network element. It can be deployed in a unified manner with the CMS and can be deployed independently. The IVM has a CMS-oriented intelligent application call and feedback interface, which can realize the management of the IVU capability and support the IVU capability. Inquire. Among them, the ability management of the IVU mainly includes: Accepting the ability of the IVU to report, save, manage, and schedule the capability information of the IVU under its jurisdiction. In addition, the IVM also supports user intelligent application policy configuration, saves user-configured smart application policies, and supports intelligent application policy query.

 Further, the composition of the IVW is as shown in FIG. 6, and mainly includes: an IVU management scheduling module 61 and an analysis result saving module 62. The IVU management scheduling module 61 is configured to save and manage the capability set of the administered IVU, select an IVU that meets the capability requirement from the administered IVR according to the request of the video monitoring system, and invoke the analysis result saving module 62, and connect the IVU management. The scheduling module 61 is configured to receive an output result of the IVU analyzing and identifying the image or video stream input from the video surveillance system to the IVU, and save the result.

 The IVW may further include: a policy management module 63, and a connection IVU management scheduling module 61, configured to provide a smart application policy configuration service, and save the configured smart application policy.

 The policy management module 63 accepts the call of the video monitoring system to the IVW saved smart application policy, and notifies the IVU management scheduling module 61; the IVU management scheduling module 61 feeds back the execution strategy according to the invoked smart application policy and the saved IVU capability set. The address of the IVU is given to the video surveillance system.

 The IVU connects to the IVM and the VTDU. It is a central service platform network element. It can be deployed in a distributed manner. It can be deployed in a unified manner with the CMS and can be deployed independently. It can realize the required information from the input video or picture and output the recognition result to the IVM; it can also directly transmit the original information of the passenger flow intelligent statistics to the IVM.

Further, the composition of the IVU is as shown in FIG. 7, and mainly includes: a capability reporting module 71 and a processing module 72. The capability reporting module 71 is configured to report the capability of the IVU to the IVW. The management and scheduling of the IVW is accepted; the processing module 72 is configured to analyze and identify the image or video stream input from the video surveillance system to the IVU, and output the processing result to the IVW.

 The processing module 72 is further configured to: when the IVU is called by the IVW as the execution policy, acquire an image or a video stream from the destination indicated by the video monitoring system, analyze and identify the input image or the video stream, and output the processing result. Give IVW.

 In the video processing system shown in FIG. 1, the IVU can report its capability set to the IVW for storage after the registration to the IVW is successful; correspondingly, when the IVW receives the request from the video surveillance system, according to its own storage The set of capabilities of the IVU, selecting the corresponding IVU that performs video analysis and recognition processing.

 The IVW can also store the smart application policy configured by the user, and provide the query and modification service of the smart application policy; when receiving the request of the video monitoring system, query the stored smart application policy, and obtain the smart application policy according to the query, and itself The set of stored IVU capabilities, select the corresponding IVU that performs video analysis and recognition processing.

 In the video processing system shown in FIG. 1, the intelligent application policy saved by the IVW is called by the CMS inside the video monitoring system, and the address of the IVU of the IVW feedback execution policy is given to the CMS, and the destination indicated by the CMS is the PU inside the video monitoring system. Then, correspondingly, the IVU obtains the video stream from the PU indicated by the CMS as:

 The IVU requests the CMS to request an image or video stream of the PU, and the CMS forwards the request to the corresponding PU;

The PU returns a response message to the CMS, and the CMS selects a VTDU for forwarding the image or video stream, and sends the response message to the IVU;

 The CMS notifies the selected VTDU of the address of the PU transmitting the image or video stream and the address of the IVU receiving the image or video stream;

 A session is established between the PU and the IVU through the VTDU, and the PU image or video stream is forwarded to the IVU via the VTDU.

The IVU obtains the video stream from the PU indicated by the CMS, and may also be: The IVU requests an image or video stream of the PU from the CMS, and the CMS forwards the request to the corresponding PU;

The PU returns a response message to the CMS, and the CMS sends a response message to the IVU;

 A session is established between the PU and the IVU, and the IVU receives the image or video stream sent by the PU.

 The video processing method implemented by the above video processing system mainly includes: IVW saves and manages the capability set of the administered IVR, and selects an IVU that satisfies the capability requirement from the administered IVR according to the request of the video monitoring system; and the IVU accepts the IVW. Management and scheduling, analyzing and identifying the image or video stream input from the video surveillance system to the IVU, and outputting the processing result to the IVW.

 The process of registering the IVU with the IVM is as follows. As shown in FIG. 2, the method mainly includes: Step 201: The IVU sends a Register message to the IVM.

 Step 202: The IVM authenticates the received registration message, and returns a registration success response to the IVU when the authentication succeeds, and step 203 is performed.

 When the authentication fails, the IVM returns a registration failure response to the IVU, and the subsequent steps are no longer performed.

 Step 203: After receiving the registration success response of the IVM, the IVU sends an option ( OPTIONS ) message to the IVM, and the above “IV IV capability set; the capability set mainly includes: the name of the intelligent algorithm, the provider, and the restriction between the intelligent algorithms. Relationships and so on.

 Step 204: The IVM feeds back the result (success or failure) of the capability set to the IVU. When the reported result is successful, the IVM saves the set of capabilities reported by the IVU.

 The following describes the process of configuring the smart application policy on the IVM through the CU. As shown in Figure 3, it mainly includes:

 Step 301: The CU where the user is located sends a login request to the CMS, where the iCU is a user interface provided to the CU.

 Step 302: When the CU logs in successfully, the CMS returns a customized intelligent application policy set to the CU.

Steps 303~304, the CU invokes the corresponding intelligent application policy configuration interface (iCU), to the CMS Send a request to read an existing smart application policy configuration.

 Step 305: The CMS sends a request to the IVM to read the CU's existing smart application policy configuration. Step 306: The IVM searches for the existing smart application policy configuration of the CU, and returns the searched smart application policy configuration to the CMS through the response.

 In step 307, the CMS returns a response to the iCU.

 Step 308: The user performs configuration and modification of the smart application policy on the iCU. After the configuration is complete, the iCU sends the configuration information to the CMS through the intelligent configuration request.

 Step 309: The CMS sends the received configuration information to the IVM.

 Step 310: The IVM saves configuration information corresponding to the smart application policy, and returns a response to the CMS.

 Step 311: The CMS returns a response to the iCU.

 Through configuration, the user can set which PUs are identified and analyzed, and process the presentation of the results, so that the CMS can control the execution and provide the required services to the users.

 The following is a specific application of a video processing method. For example, the user needs to obtain the traffic flow of each mall. The CU where the user is located may first select the relevant PU in the configuration process through the process shown in FIG. 3, and select an appropriate IVU for analysis and identification processing. Both are saved in IVW. After the configuration is completed, an application task is formed, that is, an AS; the AS can be stored in the CMS, or can be stored on a dedicated external storage device, or can be a remote application (supporting remote storage). The specific implementation process is shown in Figure 4, which mainly includes:

 Step 401: The AS initiates a content identification request. In this embodiment, the selected PUs installed in different locations of the shopping mall perform content identification to identify the traffic information at different times.

Step 402: The request is submitted to the CMS, and the CMS initiates a request to the IVM to request the information about the IVU to be provided to the CMS, including: a list, an address, and the like of the IVU supporting the video identification function. Step 403: The IVM returns information related to the IVU (list of IVUs, addresses, etc.) to the CMS. Step 404: The CMS initiates an indication to the IVU in the list, and requests the IVU to initiate a video surveillance content acquisition request to the designated PU and process the same.

 Step 405: The IVU follows the video monitoring process to obtain the video surveillance content of the specified PU. For the specific operation, refer to the process shown in Figure 5 below.

 Step 406: After obtaining the video monitoring content, the IVU identifies the video monitoring content to obtain the human traffic information.

 Steps 407~409, the IVU will return the obtained human traffic information to the CU where the user is located via the IVM and CMS; the IVU can also directly return the human traffic information to the CMS, and then return to the CU where the user is located through the CMS.

 The process of acquiring the video surveillance image by the IVU is described in detail below with reference to FIG. 5, which mainly includes: Step 501: The IVU sends an INVITE message to the CMS to request a video stream of a video source (ie, PU).

 Step 502: The CMS sends an INVITE message to the corresponding PU.

 Step 503: The PU processes the INVITE message, returns a 200 OK response to the CMS, and provides an SDP (Session Description Protocol) offer, which includes information such as a media format list supported by the PU and a PU address.

 Step 504: After receiving the 200 OK response of the PU, the CMS selects a codec mode to be used according to the media format list supported by the PU, and selects a VTDU for forwarding the video stream (also allows the PU to directly send the video stream to the IVU. There is no need to select VTDU), and the SDP offer (ie, the codec mode used in the SDP offer and the VTDU address of the forwarded video stream) is modified accordingly, and the 200 OK response is forwarded to the IVU.

 Step 505: The IVU sends an acknowledgement (ACK) message to the CMS, where the ACK message carries an SDP Answer, including the address of the IVU.

Step 506, the CMS responds according to the SDP Answer and the 200 OK in the received ACK message. The SDP offer should control the VTDU to forward the video stream between the PU and the IVU, that is, inform the VTDU to forward the source (PU address) and destination (IVU address) of the video. If the PU sends the video stream directly to the IVU, the CMS does not need to inform the VTDU without the participation of the VTDU.

 Step 507: The CMS modifies the SDP Answer in the ACK message from the IVU, and sends an ACK message to the PU. The SDP Answer includes the codec mode selected and the VTDU for forwarding the video stream (you can also let the PU send the video stream directly to the IVU, where SDP Answer does not include VTDU).

 Step 508, a session is established between the PU and the IVU. If the VTDU is required to forward the video stream, the PU establishes a session with the IVU through the VTDU. If the VTDU is not required to forward the video stream (that is, the PU directly sends the video stream to the IVU), the PU establishes a session directly with the IVU.

 The PU sends the video stream to the IVU via the VTDU; if the video stream is directly sent by the PU to the IVU, the VIDU does not need to pass.

 The session between the PU and the IVU can be established according to the Real-time Transport Protocol (RTP) or the RTP Control Protocol (PTCP, RTP Control Protocol).

 Step 509: The IVU sends a BYE message to the CMS requesting to terminate the session.

 Step 510: The CMS controls the VTDU to stop forwarding the video stream from the PU to the IVU. If the PU sends the video stream directly to the IVU, you do not need to do this.

 Step 511: The CMS sends a BYE message to the PU, and the PU stops sending the video stream.

 In step 512, the PU returns a 200 OK response to the CMS.

 In step 513, the CMS forwards the 200 OK response to the IVU.

 In summary, the present invention provides a video analysis and recognition service in a flexible and configurable manner on the basis of the existing video surveillance system; the video content of different PUs can be flexibly selected for analysis and recognition processing, based on the existing video. The monitoring system unifies the video processing services that provide video analysis and recognition capabilities and results in a flexible and configurable manner.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Scope

Claims

Claim 1. A video processing system, the system comprising: an intelligent management platform IVM and an intelligent video processing unit IVU,  The IVW is used to save and manage the capability set of the IVU under the jurisdiction, and according to the request of the video monitoring system, select an IVU that meets the capability requirement from the IVU under the jurisdiction and call it;  The IVU is configured to accept management and scheduling of the IVW, analyze and identify an image or video stream input from the video surveillance system to the IVU, and output the processing result to the IVW.  2. The video processing system according to claim 1, wherein the IVW is further configured to provide an intelligent application policy configuration service, and save the configured smart application policy.  The video processing system according to claim 2, wherein the IVW is further configured to: accept a call of the video application system to the smart application policy saved by the IVW, and feed back an IVU that executes the policy according to the saved capability set of the IVU. The address is given to the video surveillance system.  4. The video processing system of claim 3, wherein the IVU is further for acquiring an image or video stream from a destination indicated by the video surveillance system when an IVU acting as a policy is invoked by the IVW, The input image or video stream is analyzed and identified, and the processing result is output to the IVW.  5. An intelligent management platform IVW, including:  The IVU management scheduling module is configured to save and manage the capability set of the intelligent video processing unit IVU, and select an IVU that meets the capability requirement from the IVU under the control according to the request of the video monitoring system;  The analysis result saving module is configured to receive an output result of the IVU analyzing and identifying the image or video stream input to the IVU from the video monitoring system, and save the result. The IVW according to claim 5, wherein the IVW further includes: a policy management module, configured to provide a smart application policy configuration service, and save the configured smart application policy. The IVW according to claim 6, wherein the policy management module is further configured to: accept a call of the video monitoring system to the IVW saved smart application policy, and notify the IVU management scheduling module;  The IVU management scheduling module is further configured to: according to the invoked smart application policy and the saved IVU capability set, feed back an address of the IVU of the execution policy to the video monitoring system.  8. An intelligent video processing unit IVU, comprising:  The capability reporting module is configured to report the capability set of the IVU to the IVW, and accept management and scheduling of the IVW;  And a processing module, configured to analyze and identify an image or video stream input to the IVU from the video monitoring system, and output the processing result to the IVW.  9. The IVU according to claim 8, wherein the processing module is further configured to: acquire an image or video from a destination indicated by the video surveillance system when the IVU is invoked by the IVW as an execution policy; Flow, analyzing and identifying the input image or video stream, and outputting the processing result to the IVW.  10. A video processing method, the method comprising:  The intelligent management platform IVW saves and manages the capability set of the intelligent video processing unit IVU under its jurisdiction, and selects and calls the IVU that satisfies the capability requirements from the IVUs under the jurisdiction of the video surveillance system;  The IVU accepts the management and scheduling of the IVW, analyzes and identifies the image or video stream input to the IVU from the video surveillance system, and outputs the processing result to the IVW.  The video processing method according to claim 10, wherein the method further comprises: the IVW providing an intelligent application policy configuration service, and saving the configured smart application policy. The video processing method according to claim 11, wherein the method further comprises: the IVW accepting a call of the video monitoring system to the smart application policy saved by the IVW, And feeding back the address of the IVU of the execution policy to the video monitoring system according to the saved IVU capability set.  13. The video processing method according to claim 12, wherein the method further comprises: the IVU executing the policy acquiring an image or a video stream from the destination indicated by the video monitoring system, and analyzing and identifying the input image or video stream Processing, and outputting the processing result to the IVW.  The video processing method according to claim 13, wherein the intelligent application policy saved by the IVW is called by the video monitoring platform CMS inside the video monitoring system, and the IVW feeds back the address of the IVU of the execution policy to the CMS, the destination indicated by the CMS. The front end PU inside the video monitoring system;  Correspondingly, the IVU obtains an image or video stream from the PU indicated by the CMS as:  The IVU requests an image or video stream of the PU from the CMS, and the CMS forwards the request to the corresponding PU;  The PU returns a response message to the CMS, the CMS selects a video distribution unit VTDU that forwards an image or video stream, and sends the response message to the IVU;  The CMS notifies the selected VTDU of the address of the PU transmitting the image or video stream and the address of the IVU receiving the image or video stream;  A session is established between the PU and the IVU through the VTDU, and the image or video stream of the PU is forwarded to the IVU via the VTDU.  The video processing method according to claim 13, wherein the intelligent application policy saved by the IVW is called by the CMS inside the video monitoring system, and the address of the IVU of the IVW feedback execution policy is given to the CMS, and the destination indicated by the CMS is video monitoring. PU inside the system;  Correspondingly, the IVU obtains an image or video stream from the PU indicated by the CMS as: The IVU requests an image or video stream of the PU from a CMS, and the CMS forwards the request to a corresponding PU; The PU returns a response message to the CMS, and the CMS sends the response message to the IVU; A session is established between the PU and the IVU, and the IVU receives an image or video stream sent by the PU.