ImageAnalysisFilter QML Type

(BETA) Analyzes video frames received by VideoOutput. More...

Import Statement: import ArcGIS.AppFramework.DeepLearning 1.0




Detailed Description

The ImageAnalysisFilter component provides a video filter that utilizes deep learning models to analyze video frames received by the VideoOutput component.

This component is in beta. Elements are subject to change in future releases.

The following code sample analyzes video frames collected from the device's camera.

Page {
    property url modelFileUrl: AppFramework.fileInfo("~/ArcGIS/Data/leaf_classifier.tflite").url

    Camera {
        id: camera

    ImageAnalysisFilter {
        id: imageAnalysisFilter

        interval: 250

        analyzer {
            modelSource: modelFileUrl
            minimumScore: 0.5

        overlay {
            visible: true
            font {

        resultsModel {
            unique: false

        cameraOrientation: camera.orientation
        videoOrientation: videoOutput.orientation

        onDetected: {
            console.log("Detected:", JSON.stringify(result));

    VideoOutput {
        source: camera
        filters: [ imageAnalysisFilter ]
        anchors.fill: parent

Property Documentation

[read-only] analyzer : VideoAnalyzer

The VideoAnalyzer component used by this filter. This component is responsible for analyzing the video frame using the deep learning model.

cameraOrientation : double

The orientation of the camera, in degrees.

interval : int

The interval, in milliseconds, of how often the ImageAnalysisFilter will invoke the VideoAnalyzer to analyze the current video frame. Default is 100.

[read-only] overlay : ImageOverlay

Stylized text for depicting deep learning results from the VideoAnalyzer on the video frames.

[read-only] resultsModel : AnalysisResultsModel

Provides a read-only data model for the analysis results.

videoOrientation : double

The orientation of the video, in degrees.

Signal Documentation

detected(AnalysisResult result)

Signal emitted when the analysis of one video frame is finished, and the analysis results are ready.

Note: The corresponding handler is onDetected.

Method Documentation

bool classEnabled(string name)

Returns true if the specified class is enabled. If the class is not enabled, the text will not be shown on the ImageOverlay when it is the result with the highest score.

The name parameter

The name of the class to check is enabled.

See also setClassEnabled().

setClassEnabled(string name, bool enabled)

Sets the specified class to be either enabled or disabled.

The name parameter

The name of the class to enable or disable.

The enabled parameter

True if the class is to be enabled, false if the class is to be disabled.

See also classEnabled().

Your browser is no longer supported. Please upgrade your browser for the best experience. See our browser deprecation post for more details.