Class GoodFeaturesToTrack

Namespace: Bonsai.Vision

Assembly: Bonsai.Vision.dll

Represents an operator that finds strong corner features for each image in the sequence.

Editor

To open the Region of Interest (ROI) editor, start the workflow and click on the ellipsis button inside the RegionOfInterest property in the property grid, or double-click the operator node while holding Ctrl.

Create ROI

To create a new ROI, press and hold the left mouse button anywhere in the editor window and drag the mouse to define the region of interest. Release the button to complete the selection.

Note

If an ROI is already defined, pressing and holding the left mouse button outside the boundaries of the existing ROI will create a new ROI, and clear the previous one.

Move ROI

To move an ROI, press and hold the left mouse button anywhere inside the existing rectangular region and drag the mouse to reposition it. Releasing the button will complete the move.

Clear ROI

To clear any existing ROI, right-click anywhere in the editor window.

Tip

Undo and redo operations are supported on any of the above actions. To undo an action press Ctrl+Z on your keyboard. To redo something you've undone, press Ctrl+Y. You can press these shortcuts multiple times if you want to undo or redo multiple steps.

public class GoodFeaturesToTrack : Transform<IplImage, KeyPointCollection>

Inheritance: object

Combinator<IplImage, KeyPointCollection>

Transform<IplImage, KeyPointCollection>

GoodFeaturesToTrack

Inherited Members: Combinator<IplImage, KeyPointCollection>.Process(IObservable<IplImage>)

object.ToString()

object.Equals(object)

object.Equals(object, object)

object.ReferenceEquals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

Properties

MaxFeatures: Gets or sets the maximum number of corners to find.

MinDistance: Gets or sets the minimum accepted distance between detected corners.

QualityLevel: Gets or sets the minimal accepted quality for image corners.

RegionOfInterest: Gets or sets the region of interest used to find image corners. If the rectangle is empty, the whole image is used.

Methods

Process(IObservable<IplImage>): Finds strong corner features for each image in an observable sequence.

Process(IObservable<Tuple<IplImage, IplImage>>): Finds strong corner features for each image in an observable sequence, where each image is paired with a mask where zero pixels are used to indicate areas in the original image from which features should be rejected.

Table of Contents