IMAGE
- The image on which to perform the detectionDETECTED_OBJECT
- the type of object representing a detection.public interface ObjectDetector<IMAGE extends Image<?,IMAGE>,DETECTED_OBJECT>
Rectangle
representing the spatial location of the
detection, or it could be much more complex.
The interface allows for multiple detections to be returned from the input
image. If required, these detections could be filtered by a
DetectionFilter
as a post-processing operation.
This interface is implicitly for single-scale detection. If your detector
works at multiple scales, then implementing the
MultiScaleObjectDetector
would be a better idea.
Modifier and Type | Method and Description |
---|---|
List<DETECTED_OBJECT> |
detect(IMAGE image)
Detect objects in the given image and return representations of them.
|
void |
setROI(Rectangle roi)
(Optional operation).
|
List<DETECTED_OBJECT> detect(IMAGE image)
If no objects are detected, null
or an empty list may be
returned.
image
- the image to detect the object in.null
if none are found.