CN113852787A

CN113852787A - Intelligent application deployment method, device and system

Info

Publication number: CN113852787A
Application number: CN202111024548.6A
Authority: CN
Inventors: 王震宇
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2018-08-09
Filing date: 2018-08-09
Publication date: 2021-12-28
Also published as: CN110830759A; CN110830759B

Abstract

The application provides a scheme for intelligent application deployment, which is used in the field of video monitoring. In one possible approach, a deployment device receives a deployment request for a smart application, the deployment request including an identification of the smart application and an amount of resources required; in addition, the deployment device acquires analysis capability information of the analyzer and the intelligent camera, wherein the analysis capability information is resource information related to intelligent application deployment; then, according to the deployment request and the analysis capability information, a deployment device determines an object for deploying the intelligent application; finally, the deployment device sends a deployment command to the object, wherein the deployment command is used for instructing the object to deploy the one or more smart applications in the deployment request, and the deployment command comprises the identification of the one or more smart applications. The deployment of the intelligent application is carried out by cooperatively managing available resources of the camera and the analyzer, the deployment process is simplified by the intelligent application deployment scheme provided by the application, and the expandability is high.

Description

Intelligent application deployment method, device and system

Technical Field

The invention relates to the field of video monitoring, in particular to a method, a device and a system for intelligent application deployment.

Background

With the development of video monitoring technology, monitoring is developed from simple video recording to intellectualization. Such a trend of intellectualization also deeply affects the development of front-end devices (e.g., cameras), which are currently in a trend of more and more intellectualization. Meanwhile, the automatic identification accuracy of the intelligent front-end equipment exceeds that of human beings due to the rapid development of technologies such as computer vision, deep learning and the like. With the application of the intelligent front-end equipment, the monitoring efficiency is improved, the labor cost is greatly saved, and the intelligent monitoring system plays an important role in promoting modernization of safe cities, smart cities and the like, industrial production automation and the like.

Although current smart cameras can achieve partial analysis functions, the analysis capabilities of different cameras vary, and smart cameras are generally required to be used in the same monitoring system as ordinary cameras without analysis capabilities. In addition, the monitoring system management needs to comprehensively consider the existing analysis capability in the existing monitoring system. These differences present certain challenges to deploying intelligent applications. In the prior art, the method is generally adopted that before a monitoring system is used, the intelligent cameras contained in the system are fixedly configured one by one according to the self capacity of the intelligent cameras, and the complexity is high. In addition, when the task of the monitoring system changes, the system needs to be reconfigured one by one, and the flexibility is poor.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method, an apparatus, and a system for deploying an intelligent application, so as to reduce complexity of deployment of the intelligent application and improve flexibility of the intelligent application.

In a first aspect, an embodiment of the present application provides a method for deploying an intelligent application. The deployment method is used in a monitoring system, and the monitoring system comprises an analyzer and a smart camera. The deployment method comprises the following steps:

receiving a deployment request of the intelligent application, wherein the deployment request comprises an identification of the intelligent application and a required resource amount;

acquiring analysis capability information of the analyzer and the intelligent camera, wherein the analysis capability information is resource information related to intelligent application deployment;

determining at least one of the analyzer and the intelligent camera as an object for deploying the intelligent application according to the deployment request and the acquired analysis capability information;

sending a deployment command to the object, wherein the deployment command is used for instructing the object to deploy one or more smart applications in the deployment request, and the deployment command comprises the identification of the one or more smart applications.

It should be noted that the executing subject of the deployment method is a orchestrator, an intelligent orchestration center, a deployment device, a deployment apparatus, or a monitoring platform.

In one possible implementation, the required amount of resources includes one or more of a minimum amount of computing resources required, a minimum amount of memory required, and a minimum amount of memory required.

In one possible implementation, the acquiring analysis capability information of the analyzer and the smart camera includes: sending a request message for analyzing capability information to the analyzer and the intelligent camera; receiving a reply message including the analysis capability information from the analyzer and the smart camera. The analysis capability information is acquired through an active query mode, and better flexibility can be provided.

In one possible implementation, the analysis capability information includes: hardware architecture information and amount of available resources. Optionally, the analyzing capability information further includes: the method comprises the following steps of currently deployed intelligent application information, wherein the intelligent application information comprises: identification of the intelligent application and the amount of resources occupied.

In one possible implementation, the deployment request further includes: maximum resource occupancy allowed.

In a possible implementation, the determining, according to the deployment request and the acquired analysis capability information, that at least one of the analyzer and the smart camera is an object for deploying the smart application includes: and determining at least one of the analyzer and the intelligent camera as an object for deploying the intelligent application according to the deployment request, the software package information of the intelligent application and the acquired analysis capability information, wherein the software package information of the intelligent application is hardware architecture information supported by the intelligent application software version.

In a possible implementation, the determining, according to the deployment request and the acquired analysis capability information, that at least one of the analyzer and the smart camera is an object for deploying the smart application includes: deploying the smart application using the smart camera; or, deploying the smart application using the analyzer; alternatively, the smart application is co-deployed using the analyzer and the smart camera.

In one possible implementation, the method further comprises:

receiving update information for the smart application;

determining whether to refresh the deployment of the intelligent application or not according to the deployment request, the updating information and the acquired analysis capability information;

when it is determined that the deployment of the smart application is to be refreshed, sending another deployment command to the object, the another deployment command being used to instruct the object to refresh the deployment of the smart application.

Optionally, the another deployment command is to instruct the object to delete all currently deployed smart applications; or, the another deployment command includes an identification of the smart application that was deployed after the object refresh.

In a second aspect, an embodiment of the present application further provides a deployment apparatus for intelligent application. The deployment device is used in a monitoring system, which comprises an analyzer and a smart camera. The deployment device includes a processor and a transceiver. Wherein the transceiver is configured to perform the receiving and transmitting steps of the first aspect or any one of the possible implementations of the first aspect; the processor is configured to perform other steps.

In a third aspect, an embodiment of the present application further provides a monitoring system, where the monitoring system includes the deployment apparatus as described in the second aspect, and a smart camera and an analyzer.

In a fourth aspect, an embodiment of the present invention provides a computer storage medium, which includes instructions that, when executed on a computer, cause the computer to perform the method provided in the first aspect or any implementation of the first aspect.

In a sixth aspect, embodiments of the present application provide a computer program product containing instructions, which when executed on a computer, cause the computer to perform the method provided in the first aspect or any implementation of the first aspect.

Compared with the fixed configuration provided by the prior art, the technical scheme provided by the application can reduce the complexity of intelligent application deployment and improve the flexibility of the intelligent application deployment by uniformly managing the analysis resources in the monitoring system and reasonably deploying and allocating the analysis resources in combination with a specific deployment request.

Drawings

Fig. 1 is a schematic view of an application scenario in which the technical solution of the present application is applied;

fig. 2 is a schematic view of a monitoring system according to an embodiment of the present disclosure;

fig. 3 is a flowchart of a method for deploying an intelligent application according to an embodiment of the present application;

fig. 4 is a flowchart of another method for intelligent application deployment according to an embodiment of the present disclosure;

FIG. 5 is a specific example of the process flow of FIG. 4;

fig. 6 is a flowchart of another method for deploying an intelligent application according to an embodiment of the present application;

FIG. 7 is a block diagram of a possible apparatus provided in an embodiment of the present application;

fig. 8 is a block diagram of another possible device according to an embodiment of the present disclosure.

Detailed Description

The technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention. The embodiments of the present application refer to two or more. "and/or" is used to describe the association relationship of the associated objects, and means that there may be three relationships. For example, a and/or B, may represent: a exists alone, A and B exist simultaneously, and B exists alone. The specific methods of operation in the method embodiments may also be applied in the apparatus embodiments.

Currently, video monitoring is widely applied to the fields of industry, security, daily life and the like, and a video monitoring camera always pursues intellectualization. However, the analysis capabilities of the current smart cameras are different, and currently, a method for fixedly configuring the smart application deployed according to the capabilities of each smart camera has certain complexity and low expandability. In addition, the requirements of the intelligent application may change with the change of specific service requirements, which may also result in that the fixed configuration for the cameras is no longer applicable, and the configuration needs to be updated one by one, thereby having a problem of poor flexibility. Therefore, in order to solve the problems existing in the existing monitoring system, the embodiment of the present application provides a new deployment scheme of the smart application, so as to improve the flexibility of the configuration of the smart camera and simplify the configuration operation.

Fig. 1 is a schematic view of an application scenario to which the technical solution provided by the present application is applicable. As shown in fig. 1, the scenario includes a plurality of intelligent applications (10,11 and 12) and a monitoring system 100. Wherein, the monitoring system 100 further comprises a monitoring platform 101 and a front-end device (102 and 105). In particular, the front-end devices may include intelligent front-end devices of different capabilities. For example, intelligent front-end devices that support different hardware platforms (which may also be referred to as hardware architectures), such as: a Graphics Processor (GPU) or a Network Processor (NPU). As another example, the front-end devices include intelligent front-end devices of different resources. Among other things, resources may include, but are not limited to: computing resources, memory resources, and storage resources. The front-end device may also include a common front-end device. It should be noted that the difference between the common front-end device and the intelligent front-end device is as follows: the intelligent front-end device can be used for deploying intelligent application to analyze the multimedia data acquired by the intelligent application. While the common front-end device needs to process the multimedia data acquired by the common front-end device by means of the analysis capability of the monitoring platform. That is, the common front-end device needs to send the multimedia data acquired by the common front-end device to other devices, so that the other devices can analyze the data. In this application, multimedia data refers to video data, or data including video and audio. In a specific application scenario, the monitoring system may not include a common front-end device. Subsequently, the front-end equipment capable of acquiring the multimedia data is simply referred to as a camera. It should be noted that the camera mentioned in this application may be a professional camera device (for example, a camera for security protection), a non-professional device with a camera function (for example, an on-board recorder or other devices with a camera function), or any other front-end device with a function of acquiring at least video/images. Alternatively, the camera may have other capabilities, such as acquiring audio, audible and visual alerts, etc. This is not a limitation of the present application.

Figure 1 gives examples of various intelligent applications. Wherein 10 refers to face recognition application, 11 refers to license plate recognition application, and 12 refers to comprehensive application of intrusion detection and face recognition. As can be seen from these examples, a smart application may include one or more specific applications. The application does not limit the number and types of specific applications contained in an intelligent application. It should be noted that the implementation of the intelligent application needs to be assisted by a monitoring system. In one possible implementation, the smart application may be deployed through a smart camera. This approach typically requires that the analysis resources of the smart camera be able to meet the requirements of the smart application. In another possible implementation, the smart application needs to be implemented jointly by the monitoring platform and the smart camera. In particular, different specific applications may be deployed according to specific requirements of the smart application. In yet another possible implementation, the smart application is deployed only through the monitoring platform.

In the above scenario, the interface for connecting the monitoring platform and the front-end device may be a wired or wireless communication mode. The wired mode may include a Transmission Control Protocol/Internet Protocol (TCP/IP) communication technology in the ethernet technology, a User Datagram Protocol (UDP) technology or a standard Universal Serial Bus (USB) port, a COM interface, and other similar standard ports. The wireless communication mode may include WiFi, bluetooth, ZigBee, or Ultra Wideband (UWB) technology. The corresponding connection mode can be selected according to the actual application scene and the hardware form of the front-end equipment.

Fig. 2 is a schematic diagram of a monitoring system according to an embodiment of the present application. The monitoring system 200 includes a monitoring platform 201 and cameras (202a, 202b, 203a and 203 b). Specifically, the monitoring platform 201 includes an orchestrator 201a, a plurality of analyzers 201b (analyzer 1, analyzer 2, …, and analyzer n), and a software repository 201 c. Optionally, the monitoring platform may further include a video/image memory (not shown in fig. 2), an analysis result processing center (not shown in fig. 2), and the like. The orchestrator 201a is used to implement the deployment function of the intelligent application. For example, it is used to perform the method steps of fig. 3-6 with respect to the orchestrator. The orchestrator may also be referred to as an orchestration center, a deployment device, or a deployment apparatus. The name of the organizer is not limited in this application. The analyzer 201b is an analysis-capable device provided for the monitoring platform, and can be used for deploying intelligent applications. It should be noted that the analyzer 201b may be independent from the monitoring platform 201. The software repository 201c is used to store software packages of intelligent software that can be adapted to different hardware platforms. That is, an application may have multiple pieces of software available for different hardware platforms stored in a software repository. The software repository 201c may also be independent of the monitoring platform 201. It should be noted that, in addition to the software warehouse, the monitoring platform may also collect software package information of the smart application, where the software package information includes hardware architecture information supported by the version of the smart application software. It should be noted that the package information is not necessary. For example, if the default application software can support all hardware platforms used in the monitoring system, then this information is not needed. In fig. 2, the

cameras

202a and 202b are smart cameras, and 203a and 203b are general cameras, and the related description refers to the description of the front-

end device

102 and 105 in fig. 1, and will not be repeated here.

It should be noted that the monitoring platform can be implemented by combining customized software with general-purpose server hardware. For example, each component is implemented by a server and proprietary software; or all components may be implemented on a single server. Or, optionally, the monitoring platform can be implemented by deploying customized software on the cloud server. That is, the monitoring platform functionality may be deployed by renting public cloud server hardware resources and configuring proprietary software. The method and the device do not limit how the monitoring platform is actually deployed.

The following describes a deployment process for implementing the intelligent application proposed in the present application, with reference to the example of fig. 3. The step execution subject shown in fig. 3 is an orchestrator 201 a.

As shown in fig. 3, a deployment flow of an intelligent application includes the following steps:

step 301, receiving a deployment request of an intelligent application, wherein the deployment request comprises an identifier of the intelligent application and a resource amount required by the intelligent application;

specifically, the orchestrator receives a deployment request for the smart application. This deployment request may include one or more specific applications. For example, a combination of the aforementioned face recognition and license plate recognition applications may be included, or a greater number of applications. As another example, it may be a single license plate recognition application. The deployment request includes information that can uniquely identify the application. Specifically, the identifier may be identified by a character name, a number, or a uniform Resource identifier (URL), which is not limited in this application. If a smart application includes multiple specific applications, an identifier may also be provided for each specific application to indicate differentiation. In addition, the deployment request also includes the amount of resources required. In particular, the required amount of resources includes one or more of a minimum amount of computing resources, a minimum amount of memory required, and a minimum amount of memory required. More generally, the amount of resources referred to subsequently in this application includes one or more of computing resources, memory resources, and storage resources. For more contents that the deployment request may include, refer to the related descriptions of fig. 4-5, which are not described herein. It should be noted that a specific application may also be referred to as a smart application.

Step 302, acquiring analysis capability information of an analyzer and an intelligent camera, wherein the analysis capability information is resource information related to intelligent application deployment;

specifically, the orchestrator obtains analysis capability information that the monitoring system can provide. In the monitoring system, the intelligent camera has analysis capability. In addition, special analysis equipment is often provided to supplement the lack of camera analysis capability in the monitoring system. It should be noted that, the smart camera is used as a main body for acquiring multimedia data, and compared with a special analysis device, the real-time performance of analysis is better, and bandwidth resources are saved. The main reason is that special analysis equipment needs to acquire the data to be analyzed, which is usually large in data volume, consumes communication bandwidth and may require long transmission time.

The analysis capability information refers to resource information related to heel deployment of the intelligent application. The analysis capability information may also be referred to as intelligent processing capability information or intelligent analysis capability, etc. For example, for the resource amount information mentioned in step 301. The analyzer and the smart camera may provide different amounts of resources at a time. Therefore, the orchestrator needs to obtain this information. Optionally, the analysis capability information may further include priority information of the analysis capability information-equipped device. In particular, it can be implemented in various ways. In one possible implementation, this may be distinguished by giving the device identification or hardware platform information supported by the device. For example, the orchestrator may default to a smart camera having priority over the analyzer. That is, deployment of smart applications is prioritized using smart cameras. Even more, specific priority values may be set on the scheduler for different device names. In another possible implementation, the prioritizer obtains the priority information as direct priority information. For example, the analysis capability information includes a priority field indicating the priority of the body that transmits the analysis capability information. It should be noted that, if the factor of priority needs to be considered when deploying the intelligent application, the deployment request in step 301 may carry the specific requirement of priority; alternatively, the priority of each specific application may be configured in advance for the orchestrator. The former is more flexible. Those skilled in the art will appreciate that other factors (e.g., total resource information, etc.) may be considered in setting the priority information of the devices with analysis capability to distinguish between devices with different capabilities. This is not a limitation of the present application.

It should be noted that there are many specific implementations of this step. For example, the orchestrator actively acquires analysis capability information of the analyzer and the camera. The detailed description refers to the description of fig. 5, and is not repeated herein. Alternatively, the analyzer and camera report periodically or aperiodically. The first implementation mode is more flexible and has better real-time performance. However, different monitoring systems can be obtained according to actual needs, and the application does not have any limitation on the specific implementation manner.

Alternatively, the composer may selectively acquire the analysis capability information of part of the smart cameras. If the application types or other limiting conditions included in the deployment request for the smart applications are not satisfied, the orchestrator may obtain only the analysis capability information of the corresponding smart cameras capable of deploying the applications. For example, if only cameras at a specific location (e.g., a park doorway) are required to deploy face recognition, then the orchestrator may correspondingly only obtain analysis capability information for those cameras.

It should be additionally noted that the analysis capability information may also include information of the smart application already deployed on the current object (e.g., analyzer or camera). And may specifically include the identification of the smart application that has been deployed and the resources that are occupied.

In addition, it should be noted that

steps

301 and 302 are not executed in strict order. The method may be performed in a first step 301, a second step 302, or vice versa, or simultaneously. This is not a limitation of the present application.

Step 303, determining at least one of the analyzer and the smart camera as an object for deploying the smart application according to the deployment request and the acquired analysis capability information;

specifically, the orchestrator determines, according to the deployment request and the analysis capability information acquired in step 302, an object to deploy the application included in the deployment request. In the determination process, the orchestrator needs to take into account a number of factors. For example, the conditions of the constraints in the deployment request, the available resources of the analyzer and the intelligent camera, and some policy information in the local. In one possible implementation, the orchestrator selects a smart camera to deploy the application to be deployed. In another possible implementation, for example, when the analysis capability of the smart camera is insufficient, it may be considered to deploy the application to be deployed only with the analyzer; or the intelligent camera is adopted to deploy part of the application, and the analyzer is adopted to deploy the other part of the application. This is not a limitation of the present application. It should be noted that, in consideration of the fact that the analysis real-time performance of the smart camera is good, the orchestrator may set a local policy, preferentially select the smart camera to deploy the smart application, or preferentially adopt the smart camera to deploy the smart application with high priority. The advantage of doing so is that the real-time nature advantage of make full use of intelligent camera guarantees the timely processing of intelligent application and output processing result.

Step 304, sending a deployment command to the object, where the deployment command is used to instruct the object to deploy one or more smart applications in the deployment request, and the deployment command includes an identifier of the one or more smart applications.

Specifically, the orchestrator sends a deployment command to the object determined in step 303 to instruct the object to deploy the smart application formulated in the deployment command. It should be noted that the smart application identifier included in the deployment command may be a name of the smart application and/or a software storage address of the smart application. For example, the smart command may only include the name of a specific application (e.g., facedect _ arch x86_64.2.0.0.bin), and an object receiving the deployment command may go to a storage address obtained in advance to download the corresponding application for deployment (e.g., http:// registration. xxx. com/renderer /). As another example, the deployment command may include both the name of the smart application (e.g., facedect _ archX86_64.2.0.0.bin) and the downloaded server address (e.g., http:// registration. xxx. com/renderer /), and the object receiving the deployment command may combine the two information to download the corresponding application for deployment. As another example, a deployment command may only include a complete software download address (e.g., http:// registration. xxx. com/finder/facedelete _ arch x 86-64.2.0.0. bin), through which an object receiving the deployment command may directly complete the deployment of the associated application.

According to the technical scheme, the analysis resources in the monitoring system are managed in a unified mode, reasonable deployment and distribution are carried out by combining with specific deployment requests, and complexity of intelligent application deployment can be reduced.

Hereinafter, embodiments of the present application will be described in further detail based on some common aspects applicable to the present application described above. It should be noted that the terms "first," "second," and the like in this application are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments described herein are capable of operation in other sequences than described of illustrated herein.

Fig. 4 is a flowchart of another method for deploying an intelligent application according to an embodiment of the present application. In the present embodiment, the system used is the monitoring system shown in fig. 2. Specifically, the monitoring system comprises deployment equipment, K analyzers, N intelligent cameras and R ordinary cameras, wherein K and N are positive integers larger than or equal to 1, and R is a positive integer larger than or equal to zero. The deployment device may be the orchestrator shown in fig. 2. In this embodiment, the software warehouse stores software packages supporting all hardware architectures in the current monitoring system. It should be noted that the deployment apparatus may also be a deployment device. That is, the deployment function may be performed by a stand-alone device or may be performed by a portion of a device.

It should be noted that the present embodiment mainly describes a flow of steps related to the smart camera. Although a common camera cannot deploy an intelligent application, a scene embodying the technical solution of the present application may include the common camera. It should be noted that the flow chart shown in fig. 4 does not show each analyzer or camera individually, but all analyzers are taken as one object, all smart cameras are taken as one object, and actions and message contents are introduced by 1-2 examples. Those skilled in the art will appreciate that the specific resource conditions and specific interactive message content of each analyzer and smart camera need to be determined according to their own circumstances.

The process shown in fig. 4 includes the following steps:

step 401, receiving a deployment request of an intelligent application;

specifically, a deployment device receives a deployment request for a smart application. This request may be sent to the deployment device by a manager of the monitoring system through a Graphical User Interface (GUI) or an operation command or automatically triggered by the monitoring system. This step is similar to step 301 in fig. 3, and more descriptions of step 301 may be specifically referred to, and are not repeated here.

It should be noted that the deployment request of the smart application can be described by a deployment file (also referred to as a configuration file). The service scene (namely the set of the intelligent application to be deployed) is described through the configuration file, the service adjustment can be flexibly carried out by refreshing the configuration file, the deployment position of the intelligent application is further dynamically adjusted, and the requirement of the key intelligent application in one service scene is guaranteed to the maximum extent. An example deployment file is given in table 1.

TABLE 1 deployment File example

As can be seen from table 1, this example deployment request includes two specific applications, each with different information (name, resource requirements and priority, etc.). It should be noted that the computing resource generally refers to a CPU resource, and may be represented by a percentage (%) or a number between 1 and 100 is directly provided by omitting a% number. In addition, in order to ensure that accurate CPU resource requirements are provided, a plurality of values may be provided, each representing CPU resources required for deployment of a different platform. The amount of other types of resources is similar and will not be described in detail. For example, { CPU ═ 30; CPU _ ARM is 30; CPU _ X86_64 is 10, which respectively represents the amount of resources needed when not matched to the specific CPU model currently supported (e.g., ARM and X86_64 in this example), the amount of resources needed when the CPU is ARM, and the amount of resources needed when the CPU is X86_64.

The maximum allowed resource amount in table 1 refers to the maximum resource amount required by one application. When the analyzer or the smart camera is idle, more resource amount can be allocated to the application currently running, so as to improve the execution speed of the application. An advantage of providing this information is that the deployment device may prioritize objects with more available resources (e.g., exceeding the maximum amount of resources allowed) for deploying the corresponding smart application to increase the execution speed of the application. It should be noted that the specific information included in a specific application can be determined according to actual needs. For the description of the resource amount, refer to the related description of fig. 3, which is not repeated herein.

It should also be noted that the priorities in table 1 are used to indicate the real-time requirements of different intelligent applications. For example, a smart application with high real-time requirements may be set to a higher priority. Then, the orchestrator will preferentially deploy the high priority smart applications to the smart camera when deploying the smart applications. This priority may also be referred to as a real-time requirement. Alternatively, if the priority is less, the same priority can be distinguished by adding second priority information, so that the intelligent application with high second priority can be preferentially deployed on the intelligent camera.

Step 402, obtaining analysis capability information;

specifically, the deployment device acquires analysis capability information of the analyzer and the smart camera. This step is similar to step 302 in fig. 3, and more details can be found in step 302, which are not described herein again. Fig. 5 is a specific example of the process flow of fig. 4. In fig. 5, the deployment device obtains the analysis capability information by means of active query. Specifically, the deployment apparatus sends request information (which may also be referred to as a query request) of the analysis capability to the plurality of analyzers and the plurality of smart cameras through step 502a and step 503a, respectively. Correspondingly, the analyzer and the smart camera receiving the request send reply information of the analysis capability to the deployment device (step 502b and step 503b in fig. 5).

An example of the content included in the reply message for analysis capability is given in table 2. It should be noted that, the currently supported application information, hardware architecture, total amount of resources, and other information in the information may be carried in the reply message according to specific needs. For example, because the hardware architecture is typically unchanged for a period of time, it may be reported only once, which may save bandwidth. Only when a change occurs again does a re-reporting be required. Other resources, such as the total amount of resources, may be processed similarly, and are not described in detail.

Table 2 examples of information included in reply messages for analytic capabilities

Step 403, determining a deployment object of the intelligent application;

specifically, the deployment device integrates the information obtained in steps 401 and 402, and determines which objects the request obtained in step 401 needs to be deployed by. This step is similar to step 303 of fig. 3, and more details can be found in step 303, which are not described herein again. For example, in the present embodiment, the smart cameras 1 to 7 may deploy smart applications to be deployed; the smart cameras 8 to N-10 need to deploy a part of applications to be deployed on the analyzer and place the other part of applications to be deployed on the smart cameras for deployment due to insufficient resources of the smart cameras; the cameras N-9 to N have no resources to deploy more applications, and need to be fully assisted by the analyzer.

Step 404, send deployment command.

Specifically, the deployment device sends deployment commands to the objects determined in step 403 one by one. As shown in fig. 5 (step 505 and step 506), different content deployment commands can be sent to the analyzer and the smart camera, respectively. Taking the example given in step 403 as an example, then the deployment device needs to send deployment commands to the cameras 1 to 7 to cause these cameras to deploy all the specific applications to be deployed. The deployment device needs to send deployment commands to the cameras 8 to N-10 to cause these cameras to deploy specific ones of the smart applications. In addition, the deployment device may send a deployment command to the analyzer to deploy the remaining applications that cannot be deployed in the camera. The deployment device need not send deployment commands to the cameras 8 to N-10, but only to the analyzer, to deploy all the specific applications. It should be noted that if there are multiple analyzers, the second and third scenarios related to analyzers may need to send deployment commands to different analyzers.

It should be noted that step 501 and step 504 in fig. 5 are similar to step 401 and step 403 in fig. 4, and are not described again here.

According to the technical scheme, the analysis resources in the monitoring system are managed in a unified mode, reasonable deployment and distribution are carried out by combining with specific deployment requests, automatic deployment of intelligent application is achieved, and complexity of intelligent application deployment can be reduced. In addition, the probability of deployment errors can be reduced, for example, intelligent applications are repeatedly configured, intelligent applications are omitted to be configured, and the like.

Fig. 6 is a flowchart of another method for deploying an intelligent application according to an embodiment of the present application. In this embodiment, similar to the embodiment shown in fig. 4, the system used is the monitoring system shown in fig. 2. It should be noted that the flowchart shown in fig. 6 focuses on the steps of deploying refresh by the application. The method comprises the following specific steps:

step 600(S600), deploying the smart application;

in particular, this step may be replaced by the deployment step shown in fig. 3 or fig. 4 or fig. 5. For details, reference is made to the description of the corresponding drawings, which are not repeated herein. It should be noted that, in this embodiment, the deployment device further needs to acquire hardware platform information supported by the intelligent application stored in the software repository, and use this information as a consideration condition for selecting a deployment object. For example, if a software version supporting the GPU hardware platform is required for license plate recognition of this application, the deployment device does not consider cameras or analyzers that do not support the GPU hardware platform in the selection of the deployment object. In short, it can be understood that determining at least one of the analyzer and the smart camera as an object for deploying the smart application according to the deployment request and the acquired analysis capability information includes: and determining at least one of the analyzer and the intelligent camera as an object for deploying the intelligent application according to the deployment request, the software package information of the intelligent application and the acquired analysis capability information.

Step 601 (S601): an update request for the smart application;

specifically, the deployment device receives an update request of the smart application. For example, because of a change in the scenario, the deployment device may obtain a request for an update, and the source of the specific request may be the sending subject of the deployment request message of the smart application as mentioned in step 401. It should be noted that the subject matter of the two requests may be different. For example, the deployment request may come from system management software, while the refresh request may be platform-self policy-triggered.

As a specific example, the priority requirements in an existing intelligent application request may be updated. For example, the priority information of the two applications in table 1 is exchanged. Alternatively, it may be to refresh the content contained in the deployment request as mentioned in FIG. 3 to the smart application. Alternatively, it may be a software package that refreshes the smart application. In the subsequent steps, the real-time interchange of the two applications in table 1 is illustrated as an example. Those skilled in the art will appreciate that if there are other attribute updates, the process is similar. It should be noted that, in step 600, the deployment result for a certain smart camera (hereinafter, referred to as smart camera X) is: face recognition is deployed on the analyzer, while intrusion detection is deployed on the smart camera.

Step 602 (S602): determining whether a deployment of the smart application needs to be refreshed;

specifically, the deployment device comprehensively determines whether the current deployment needs to be refreshed or not according to the received deployment request of the intelligent application, the corresponding update request and the acquired analysis capability information. If it is determined that a flush deployment is not needed, then the current flow ends. If it is determined that a refresh deployment is required, then step 603 is performed.

In connection with the example of step 601, in the current step, because the resources of the smart camera X are limited, the deployment device determines that the smart application deployed by the deployment device needs to be refreshed by synthesizing the deployment request and the update request. Specifically, the smart camera needs to deploy face recognition, while intrusion detection needs to be deployed on the analyzer.

Step 603 (S603): upon determining to refresh the deployment of the smart application, another deployment command is sent.

Specifically, the deployment device sends another deployment command to the object that needs to be refreshed for refreshing deployment of the intelligent application. Depending on the type of refresh, another deployment command may not include the identification of any smart application, such that the object receiving the command deletes all of its currently deployed applications. Or, another deployment command is all of the smart applications that should be deployed (including already deployed and newly added), the received object can determine which ones are already present and which ones need to be newly added. Or, the command only comprises the identification of the intelligent application needing to be changed, and the command is distinguished by an indication field to be deleted or added or parameters refreshed.

In conjunction with the example of step 601 and step 602, in the current step, the deployment device needs to send another deployment command to the smart camera X to instruct the smart camera X to deploy face recognition. After receiving the other deployment command, the smart camera X determines that the intelligence that needs to be deployed should be face recognition, and obtains the software download address of the application. Then, the smart camera X deletes the current intrusion detection deployment, downloads the intelligent application software for face recognition, installs and runs, and completes the deployment of the application. Similarly, the deployment device sends a deployment command to the analyzer instructing it to deploy intrusion detection. It should be noted that intrusion detection may be deployed to an analyzer that previously deployed a face detection application, or to another analyzer. It should be noted that in either case, the face recognition application originally deployed on the analyzer needs to be deleted.

According to the technical scheme, the analysis resources in the monitoring system are managed in a unified mode, reasonable deployment and distribution are carried out by combining with specific deployment requests, and complexity of intelligent application deployment can be reduced. In addition, according to the technical scheme, the deployment object of the intelligent application can be dynamically adjusted according to the updating of the deployment request, and the deployment efficiency of the intelligent application is improved.

Fig. 7 is a diagram of a possible apparatus structure provided in the embodiment of the present application. The device 700 includes a processor 701 and a transceiver 702. Wherein the processor 701 and the transceiver 702 are coupled. The apparatus 700 is the orchestrator of fig. 3, or the deployment apparatus of fig. 4-6. Correspondingly, the device implements all the steps of fig. 3. Wherein the transceiver 702 performs the transmitting and receiving steps and the processor 701 performs the other steps. Similarly, the device implements all the steps performed by the deployment device of fig. 4-6. Wherein the transceiver 702 performs the transmitting and receiving steps and the processor 701 performs the other steps.

Fig. 8 is a block diagram of another possible device according to an embodiment of the present disclosure. The apparatus 800 includes a receiving module 801, a determining module 802, and a transmitting module 803. The apparatus 800 is an orchestrator device in fig. 3, or a deployment device in fig. 4-6. Correspondingly, the device implements all the steps of fig. 3. The receiving module 801 executes the

steps

301 and 302, the determining module 802 executes the step 303, and the sending module executes the step 804. Similarly, the device implements all the steps performed by the deployment device of fig. 4-6. Wherein, the receiving module 801 implements receiving or obtains related actions from the outside, while the sending module 803 implements sending related actions, and the determining module performs other steps (e.g., step S403 in fig. 4, step S603 in fig. 6).

While the invention has been described in connection with various embodiments, other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a review of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the word "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

The Processor or the determining module may be a Central Processing Unit (CPU), a general purpose Processor, an Artificial Intelligence (AI) chip, a Graphics Processing Unit (GPU), a Digital Signal Processor (DSP), an Application-Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), other Programmable logic devices (Programmable Gate Array), transistor logic devices, hardware components, or any combination thereof. The processor may also be a combination implementing computational and/or control functions, e.g., comprising one or more microprocessors, a combination of DSPs and microprocessors, or a combination of a CPU and a GPU or AI chip, etc.

Those skilled in the art will appreciate that all or part of the steps for implementing the above embodiments may be implemented by a program instructing associated hardware to perform the steps. The program may be stored in a computer-readable storage medium. The storage medium may be, but is not limited to, a read-only memory (ROM) or other type of static storage device that may store static information and instructions, a Random Access Memory (RAM) or other type of dynamic storage device that may store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, optical disk storage (including compact disc, laser disc, optical disc, digital versatile disc, blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application. When implemented in software, the method steps described in the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, causes, in whole or in part, the flow or functions described in accordance with any of figures 3 and 4-8. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in or transmitted from a computer-readable storage medium to another computer-readable storage medium, e.g., from a website, computer, server, or data center, over a wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL), or wireless (e.g., infrared, wireless, microwave, etc.) connection to another website, computer, server, or data center. SSD)), etc.

While the invention has been described in conjunction with specific features and embodiments thereof, it will be evident that various modifications and combinations can be made thereto without departing from the spirit and scope of the invention. Accordingly, the specification and figures are merely exemplary of the invention as defined in the appended claims and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of the invention. It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims

1. A deployment method of intelligent applications, wherein a monitoring system comprises a monitoring platform and an intelligent camera, and the deployment method comprises the following steps:

acquiring a deployment request for deploying the intelligent application;

acquiring analysis capability information of the monitoring platform and analysis capability information of the intelligent camera;

determining at least one of the monitoring platform and the intelligent camera as an object for deploying the intelligent application according to the deployment request and the acquired analysis capability information, wherein the resource of the object meets the resource amount required by the intelligent application;

sending a deployment command to the object, wherein the deployment command is used for instructing the object to deploy one or more intelligent applications in the deployment request.

2. The method of claim 1, wherein the smart application is deployed preferentially using the monitoring platform.

3. The method of claim 1, wherein the smart application is deployed preferentially using the smart camera.

4. The method of claim 1, wherein the deployment request includes an amount of resources required by the smart application.

5. The deployment method of claim 4 wherein the amount of resources required comprises one or more of a minimum amount of computing resources required, a minimum amount of memory required, and a minimum amount of memory required.

6. The method of any one of claims 1-5, further comprising: and acquiring the resource quantity required by the intelligent application from the deployment request, and determining whether the intelligent application is satisfied according to the resource quantity required by the intelligent application and the acquired analysis capability information.

7. The method of claim 6, wherein analyzing the capability information comprises: hardware architecture information and amount of available resources.

8. The method of claim 7, wherein analyzing the capability information further comprises: the method comprises the following steps of currently deployed intelligent application information, wherein the intelligent application information comprises: identification of the intelligent application and the amount of resources occupied.

9. The method of any of claims 1-6, wherein the deployment request comprises: maximum resource occupancy allowed.

10. The method of claim 6, wherein the determining, according to the obtained analysis capability information, that at least one of the monitoring platform and the smart camera is an object for deploying the smart application comprises:

and determining at least one of the monitoring platform and the intelligent camera as an object for deploying the intelligent application according to the deployment request, the software package information of the intelligent application and the acquired analysis capability information, wherein the software package information of the intelligent application is hardware architecture information supported by the intelligent application software version.

11. The method of claim 6, wherein the determining at least one of the monitoring platform and the smart camera as an object for deploying the smart application according to the obtained analysis capability information comprises:

deploying the smart application using the smart camera; or, deploying the smart application using the monitoring platform; or jointly deploying the intelligent application by using the monitoring platform and the intelligent camera.

12. The method of any of claim 6, further comprising:

receiving update information for the smart application;

13. The method of claim 12, wherein the another deployment command is instructing the object to delete all currently deployed smart applications; or, the another deployment command includes an identification of the smart application that was deployed after the object refresh.

14. The method of claim 1, wherein the deployment command includes an identification of the one or more smart applications for the object to download the smart applications for deployment.

15. The method of claim 1, wherein the monitoring platform is a server or a cloud server.

16. A deployment apparatus for smart applications, wherein a monitoring system comprises a monitoring platform and a smart camera, the deployment apparatus comprising a processor and a transceiver, wherein:

the transceiver is used for acquiring a deployment request for deploying the intelligent application;

the transceiver is further used for acquiring analysis capability information of the monitoring platform and analysis capability information of the intelligent camera;

the processor is configured to determine, according to the deployment request and the acquired analysis capability information, that at least one of the monitoring platform and the smart camera is an object for deploying the smart application, where resources of the object satisfy a resource amount required by the smart application;

the transceiver is further configured to send a deployment command to the object, where the deployment command is used to instruct the object to deploy the one or more smart applications in the deployment request.

17. The deployment apparatus of claim 16 wherein the smart application is deployed preferentially using the monitoring platform.

18. The deployment apparatus in accordance with claim 16 wherein the smart application is deployed preferentially using the smart camera.

19. The deployment apparatus of claim 16 wherein the required amount of resources comprises one or more of a minimum amount of computing resources required, a minimum amount of memory required, and a minimum amount of memory required.

20. The deployment apparatus of claim 16 wherein the transceiver, further configured to receive analysis capability information of the monitoring platform and the smart camera, comprises:

the transceiver is used for sending a request message for analyzing the capability information to the monitoring platform and the intelligent camera;

the transceiver is further configured to receive a reply message including the analysis capability information from the monitoring platform and the smart camera;

and the processor is used for determining whether the intelligent application is satisfied according to the resource amount required by the intelligent application and the acquired analysis capability information.

21. The deployment device of claim 16 wherein the analysis capability information comprises: hardware architecture information and amount of available resources.

22. The deployment device of claim 21 wherein the analysis capability information further comprises: currently supported smart application information, wherein the smart application information comprises: identification of the intelligent application and the amount of resources occupied.

23. The deployment apparatus of any of claims 16-22 wherein the deployment request further comprises: the maximum amount of resources allowed.

24. The deployment apparatus according to any of claims 16-22, wherein the determining, according to the deployment request and the acquired analysis capability information, that at least one of the monitoring platform and the smart camera is an object for deploying the smart application comprises:

and determining at least one of the monitoring platform and the intelligent camera as the intelligent application object to be deployed according to the deployment request, the software package information of the intelligent application and the acquired analysis capability information, wherein the software package information of the intelligent application is the hardware architecture information supported by the intelligent application software version.

25. The deployment apparatus according to any of claims 16-22, wherein the determining, according to the deployment request and the obtained analysis capability information, that at least one of the monitoring platform and the smart camera is an object for deploying the smart application comprises:

26. The deployment apparatus of any of claims 16-22 wherein:

the transceiver further configured to receive update information for the smart application;

the processor is further configured to determine whether to refresh the deployment of the intelligent application according to the deployment request, the update information, and the acquired analysis capability information;

when it is determined to refresh the deployment of the smart application, the transceiver is further configured to send another deployment command to the object, the other deployment command being used to instruct the object to refresh the deployment of the smart application.

27. The deployment apparatus of claim 26 wherein the another deployment command is an instruction to the object to delete all currently deployed smart applications; or, the another deployment command includes an identification of the smart application that was deployed after the object refresh.

28. The deployment apparatus of claim 16 wherein the deployment command comprises an identification of the one or more smart applications for the object to download the smart application for deployment.

29. The deployment apparatus of claim 16 wherein the monitoring platform is a server hardware or a cloud server.

30. A monitoring system comprising the deployment apparatus of any of claims 16-27, and a smart camera and a monitoring platform.

31. A deployment apparatus for smart applications, comprising:

a module for acquiring a deployment request for deploying the intelligent application;

the module is used for acquiring analysis capability information of a monitoring platform and analysis capability information of an intelligent camera, wherein the monitoring platform and the intelligent camera form a monitoring system;

determining at least one of the monitoring platform and the intelligent camera as a module for deploying an object of the intelligent application according to the deployment request and the acquired analysis capability information, wherein the resource of the object meets the resource amount required by the intelligent application;

a module that sends a deployment command to the object, the deployment command being used to instruct the object to deploy the one or more smart applications in the deployment request.

32. The deployment device of any of claims 31, wherein the device further comprises:

and confirming whether the intelligent application module is met or not according to the deployment request including the intelligent application identifier, the intelligent application required resource amount, the acquired analysis capability information of the monitoring platform and the acquired analysis capability information of the intelligent camera.