Microsoft Computer Vision -Analyze Image

The first activity in the Microsoft Computer Vision package is Analyze Image.

AnalyzeImage

This activity can analyze an image and generate a human-readable sentence that describes its contents. The algorithm actually returns several descriptions based on different visual features, and each description is given a confidence score. The final output is a list of descriptions ordered from highest to lowest confidence.

Important Point: Click on the following images to get full screen view.

How to use the activity

Drag a MicrosoftVisionScope dropzone activity from the MicrosoftComputerVision package and drop it on the canvas. Enter the Service URL and Subscription Key in the Property window.

DisplayName will be auto populated for all activities.

Note: Append the version of Microsoft Vision Scope activity in the Service URL.

Figure 1

 

     22

 

 Drag and drop an AnalyzeImage activity inside the MicrosoftVisionScope dropzone.

Figure 2

2

Figure 2.1

Specify the local ImagePath where the image is located, to analyze the image. Display Name and Version will be auto populated and will state the activity's name and version number, respectively. Create a variable under the Output header in the Result field to be passed as input string for the Input field of the write log activity.

  2

Figure 3

Drag and drop a WriteLog activity from the NotificationAutomation package. Specify the InputString and the LogLevel.

  4

Figure 4

Note- Provide the output Result value as the input string value in the Write log. And choose the log level as per requirement.

 

rs

Figure 5

The provided image is as follows:

car

Figure 6

Execute the workflow to extract comprehensive details about the aforementioned image.

In the Output window, you will see all details, such as color, image type, description of the entity in the image, etc.

   op1            car op

      Figure 6.1                                                       Figure 6.2                                  

In the event of analyzing a human image, the output will vary accordingly. It names the person (Famous and well known), their age and gender.

 

        CHURCHILL                  op              w  op

         Figure 7                                                                 Figure 7.1                         Figure 7.2                                                                             

Release date
Version
V5.0.1