Procedure

1. Log in to the Video Analysis Service management console.
   • Choose Overview and click create a job in the Quick Start area.
   • Choose Service and click Create Job in the Operation column of the desired service on the Market Subscription page.
   • Choose Jobs and click Add.
2. Set the parameters on the Create Visual Analysis Job page that is displayed, as described in Table 1.

   Table 1 Setting visual analysis service job parameters

   Parameter	Description	Example Value
   Name	Specify a job name.	Work02
   Description	Customize memorable and distinguishable job description.	N/A
   Job Type	Job Type can be Normal or Scheduled.	Normal
   Service Category	Select a visual analysis service.	Video Content Recognition
   Function	If you create this job through the Service page, Function and Version are automatically set to the latest. If you create this job through the Jobs page, you need to set Function and Version. When you select a service function, the system prompts you to authorize the service. The purpose is to allow the job to read video data from the OBS bucket and path you specified, and save the result file to the OBS bucket and directory you specified after the job is successfully executed. When you authorize the service, if the message "The number or agencies has reached the maximum." is displayed, you need to delete the agency or create an agency on the IAM management console.	Careless Sorting Recognition
   Version	Select a version.	1.0
   Edge Stream Pool	If the function you selected for Function requires an edge stream pool, you need to set Edge Stream Pool. For details about edge stream pools, see Edge Stream Pool.	N/A
   Visual Service Package	Select the visual service package you purchased. The total number of video channels for creating jobs cannot exceed the number of video channels corresponding to the capability package. If an algorithm service requires a visual capability package (for example, edge transportation intelligent algorithm capability package), you need to purchase the package and provides the package name and order number when you use the service to deliver a job. For details about how to purchase a visual capability package, contact Huawei technical support.	N/A
   Algorithm Template Settings	The parameters for configuring the algorithm template vary depending on the function. For details, see Algorithm Template Settings.	N/A
   Input Data	Select the path to the video data to be processed. The options are as follows. For details, see Step 3. OBS: When video data is stored on OBS, you only need to select a bucket and a path. In the Operation column, select some or all video files. VIS: When real-time data is accessed through VIS, you need to enter the name of the VIS data stream. Edge Camera: Read data from a specified edge camera. URL: Read data from a specified URL. Currently, only the URL of OBS is supported. You need to grant the anonymous users the permission to read the URL.For details, see Bucket ACLs and Object ACLs. Edge Restful: Read data from the custom streaming media server. VCN: Read data from VCN. NOTE: The available input data path sources are related to the algorithm service. For details, see the related documents of each algorithm service.	OBS
   Output Type	Select the path for storing the result files generated after the videos are processed. The options are as follows: For details, see Step 4. OBS: To export the results to OBS, you can select an existing bucket and path or create a path. Select an existing path: After selecting an existing bucket and path, you can obtain the job results file. Create a new path: Select the corresponding storage address and enter the storage path to obtain the job result file. DIS: Output the result data to the specified channel of DIS. DMS: Not supported. Webhook: Send the job result directly to the terminal address Webhook URL set by the user through a post request. Local Path: Save the job result on an edge node running Linux. The path must be a Linux path, for example, /opt/cloud/. This path is in accordance with the Mount Source Path set by the user when creating the job. Hosting: Store the result data on OBS, with the OBS bucket and path specified by the service owner. For details, see Obtaining the Job Result.	OBS

3. Version, Input Data, and Output Type are matched by default. You can also manually set Input Data to either of the following values as needed:
   1. OBS: When you select this value, you only need to select the OBS bucket and path and then click Select or Select All in the Operation column to select one video file or all video files on this page.
      Figure 1 Setting Input Data to OBS
      
   2. URL: Read video data from the URL of OBS, for example, https://obs-vcm.obs.cn-north-1.myhuaweicloud.com/videos/est_h264.mp4.
      Figure 2 Setting Input Data to URL
      
   3. Edge Camera: When you select this value, you need to select the edge node and edge camera, and then click Select in the Operation column or Select All under Available Data Source to select one camera or all cameras on this page.
      Figure 3 Setting Input Data to Edge Camera
      
   4. VCN: When you select this value, you need to set IP Address, Port Number, Username, Password, and Camera ID.
      Figure 4 Setting Input Data to VCN
      
4. Set Output Type to either of the following values:
   1. OBS: When you select this value, you can use either of the following two methods to output the job result file:
      • Select an existing: Select an existing bucket and path to obtain the job result file.
      • Create a new path: Select the corresponding storage address and enter the storage path to obtain the job result file.
        Figure 5 Setting Output Type to OBS
        
   2. DIS: When you select this value, you need to specify a stream name for the output result.
      Figure 6 Setting Output Type to DIS
      
   3. Hosting: When you select this value, you need to view the output result in the Algorithm Result Output column on the job details page. For details, see Obtaining the Job Result.
   4. Local Path: You can select this value only when creating an edge job. When you select this value, the output result data will be stored on the edge node running Linux. You need to enter the local path of the edge node for storing the result data. The path must be a Linux path, for example, /opt/cloud/.
      Figure 7 Setting Output Type to Local Path
      
   5. Webhook: You can select this value only when creating an edge job. When you select this value, the service sends the job result to the endpoint you preconfigured through a post request. You need to set URL and Headers.
      Figure 8 Setting Output Type to Webhook
      
5. After the configuration is complete, click Create.
   The Jobs page is displayed. You can click the name of the created job to view its details.

   Figure 9 Job list