Description:

The Describe Image step enables a FlowWright workflow to generate a description of an image in human-readable language using complete sentences.

The step is part of the Azure Cognitive Services category and is exposed in FlowWright with the display name Describe image. It accepts an image file path as input and stores the generated description in a workflow variable or global variable.

This capability allows image content to become part of an automated workflow. Instead of requiring a user to manually describe an uploaded image, the workflow can generate descriptive text that can subsequently be included in reports, documents, notifications, records, or other workflow outputs.

Typical uses include:

• Generating descriptions for uploaded images
• Adding descriptive text to business documents
• Enriching records with image descriptions
• Supporting image-based reporting
• Generating descriptive content for notifications
• Processing photographs as part of business workflows
• Supporting image-content classification and review
• Creating human-readable image information for downstream processing

The step provides a straightforward way to incorporate image description into FlowWright workflow automation.

Usage:

The Describe Image step is typically used after a workflow has received or generated an image that needs to be described.

During execution:

1. Provide the path to the image file.
2. Execute the Describe Image step.
3. The image is processed to generate a human-readable description.
4. The resulting description is stored in the configured workflow variable or global variable.
5. Continue workflow processing through the True or False path.

A typical workflow might look like this:

Typical workflow scenarios include:

• Image documentation
• Product catalog enrichment
• Automated report generation
• Document preparation
• Image-based business record creation
• Content generation from uploaded images
• Image review workflows
• Automated notification generation

To use this step, you need to set up an Azure ML service in the FlowWright application. Go to the Status > Settings > Configuration page. Select the Azure ML category from the drop-down menu. Click here to learn more about the Azure ML and Cognitive Service subscription. 

A sample Azure ML configuration is provided here for reference. 

Example:

Let’s build and execute the “clsAzureMLDescribeImageDef” example.

• Create a new process definition named “clsAzureMLDescribeImageDef” and open it in designer mode. 
• Drag a "clsAzureMLDescribeImage" step to the canvas.
• Connect the dots between the “Start” and “clsAzureMLDescribeImage” steps, as shown above.
• Define a variable or a global to store the image file path and the description text.
• Click the “updateVariable” step to configure its “Required” properties. Enter a step name. Then click the Save button. Note: Click the "AI Predict" button to have the Copilot add new process steps that match your process description. 
   

• Click the “updateVariable” step to configure its “Optional” properties. Click the button to update multiple variables. A pop-up window appears for configuration. Click the Add Row (+) button to insert an empty row. Enter a variable or global reference to store the image file path on the application server. Click the Save button. You can insert multiple variables using the Add Row button. 

• Click the “clsAzureMLDescribeImage” step to configure its “Required” properties. Provide a step name. Provide a variable or global reference for the image file path. Provide a variable or global reference to store the image description as text. Click the Save button. Note: Click the "AI Predict" button for the Copilot to add new process steps that match your process description. 
   

• The image file path indicates where the image is stored on the server. The example image used is included below for reference. 

• The “Logging” configuration is necessary for documentation and also measures workflow progress and percent complete. This is achieved by configuring the step state and percent fields individually, as shown below. Configure the “Logging” using the following properties.

• Save the process definition, create a new instance, and run it. Render the process instance. Click the process step to view its properties. The step should describe an image as text using Azure Machine Learning (ML) services, and store the result in “variable.description” as configured. 

Tips:

• Ensure that the configured Image file path points to an accessible image before executing the step.
• Store the generated description in a workflow variable when it is needed within the current workflow.
• Use a global variable when the description needs to be accessed more broadly within the workflow.
• Consider generating the image description before creating reports, documents, or notifications that require descriptive image information.
• Use the False path to handle image-processing failures or unexpected input conditions.
• Test the step with representative images before designing downstream logic that depends on the generated description.
• Combine this step with file management, reporting, document generation, notification, or database steps to create end-to-end image-processing workflows.

Notes:

• The step belongs to the Azure Cognitive Services category. 
• Its display name is Describe image. 
• An image file path is required. 
• Variable/Global to store the description is required. 
• Both inputs are defined as string properties. 
• The step has two input connections and two output connections according to its XML definition. 
• The step provides True and False execution paths. 
• The XML does not specify additional configuration properties beyond the image path and output variable/global.

Brand Detection vs. Describe Image:

Both Describe Image (clsAzureMLDescribeImage) and Brand Detection (clsAzureMLGetBrands) are Azure Cognitive Services image-analysis steps, but they focus on different types of information.

When to use each step

Use Describe Image when the workflow needs to answer:

“What does this image show?”

The resulting human-readable description can be used as workflow data for reports, documents, notifications, or other business processes.

Use Brand Detection when the workflow needs to answer:

“What popular brand information can be detected in this image?”

The resulting brand information can then be used for brand-specific processing or classification.

Definition Sample:

You may download the sample workflow definition from the link provided and import it into your FlowWright environment.

Note: Verify and complete any missing configuration after importing the sample, including:

• Image file path
• Variable/Global to store the description
• Image availability and accessibility
• Success and failure workflow branches
• Any downstream processing that consumes the generated description

After verifying the configuration, save and publish the workflow before execution.

Click here to download the sample file.