Matching compound

A compound tool that contains a key point detector and a matcher. Tools derived from this generic implementation specify which key point detector and matcher are actually used. All derived tools can be used in the matcher builder to create key point based matchers.

Inputs

• image: The image in which objects are to be detected. Different key point detection techniques require different input image types: Boundary-based detectors only work with blob (binary or labeled) images. Texture or color based detectors work with gray-level or color input.

• frame: The location of a region-of-interest for the matcher. A coordinate frame in world coordinates.

• size: The size of the region-of-interest in frame.

• classNames: A N-by-1 table containing the names of the model classes.

Outputs

• frame: A coordinate frame for each detected object. Each frame is a 4-by-4 matrix totaling in a 4N-by-4 matrix.

• size: The size (width, height) of each detected object in the corresponding frame. An N-by-2 matrix.

• modelIndex: The index of the model that matched. An N-by-1 matrix.

• classIndex: The class index of the matching model. An N-by-1 matrix.

• className: The class name of the matching model. An N-by-1 table.

• confidence: Confidence of the match (0…1). The bigger the value, the better the match. An N-by-1 matrix.