Edit

Share via

Use the object detection component in Power Apps

  • Article

The object detection component takes a photo or loads an image file to do an object detection scan. On a mobile device, the user chooses between taking a photo or selecting one already available in the device user interface. When an image is selected, the component automatically scans it to identify objects.

Note

For information about canvas apps, see What are canvas apps in Power Apps?.

Prerequisites

To start, you need a trained and published AI Builder object detection model. You bind that model to the component using AI Model in the Properties panel. When you add it to the screen, the component automatically opens the AI models pane. There you select a model from the models published in that environment.

The component is initialized after an AI model is bound to it.

Properties

Key properties

  • ModelId (AI model in the properties panel): AI model information to which the component is bound.

  • OriginalImage: The original image before processing.

  • GroupedResults: The details of detected objects. For each object, these properties are available:

    • TagId The ID of the object detected.

    • TagName The name of the object detected.

    • ObjectCount The number of times the object is present on the image.

  • Results: Contains all the outputs returned by the model.

    • BoundingBox The coordinates on the image of the object detected.

    • Confidence How confident the model is in its prediction of the object detected.

    • TagId The ID of the object detected.

    • TagName The name of the object detected.

Note

Some of these property names changed with the April 2020 updates. If you aren't seeing these properties in your app, you'll have to manually update the property names. More information: Property name changes in AI Builder components for canvas apps

Additional properties

  • Text: Text that appears on the button that activates the object detector.

  • ImageDisplayed (Show image in the properties panel): Determines whether the component displays the image.

  • DisplayMode:

    • Edit: Allows user input.

    • View: Only displays data.

    • Disabled is disabled.

  • Height: The height of the component.

  • Visible: Whether the component appears or is hidden.

  • Width: The width of the component.

  • X: The distance between the left edge of the component and the left edge of its parent container or screen.

  • Y: The distance between the top edge of the component and the top edge of the parent container or screen.

Additional design properties are available in the Advanced panel.

Accessibility guidelines

These guidelines for the Power Apps button control also apply to the form processor component.