Package org.opencv.dnn

Class DetectionModel

java.lang.Object

org.opencv.dnn.Model

org.opencv.dnn.DetectionModel

public class DetectionModel
extends Model

This class represents high-level API for object detection networks. DetectionModel allows to set params for preprocessing input image. DetectionModel creates net from file with trained weights and config, sets preprocessing input, runs forward pass and return result detections. For DetectionModel SSD, Faster R-CNN, YOLO topologies are supported.

Field Summary

Fields inherited from class org.opencv.dnn.Model

nativeObj

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	DetectionModel(long addr)
	DetectionModel(String model)	Create detection model from network represented in one of the supported formats.
	DetectionModel(String model, String config)	Create detection model from network represented in one of the supported formats.
	DetectionModel(Net network)	Create model from deep learning network.

Method Summary

Modifier and Type	Method	Description
static DetectionModel	__fromPtr__(long addr)
void	detect(Mat frame, MatOfInt classIds, MatOfFloat confidences, MatOfRect boxes)	Given the input frame, create input blob, run net and return result detections.
void	detect(Mat frame, MatOfInt classIds, MatOfFloat confidences, MatOfRect boxes, float confThreshold)	Given the input frame, create input blob, run net and return result detections.
void	detect(Mat frame, MatOfInt classIds, MatOfFloat confidences, MatOfRect boxes, float confThreshold, float nmsThreshold)	Given the input frame, create input blob, run net and return result detections.
protected void	finalize()
boolean	getNmsAcrossClasses()	Getter for nmsAcrossClasses.
DetectionModel	setNmsAcrossClasses(boolean value)	nmsAcrossClasses defaults to false, such that when non max suppression is used during the detect() function, it will do so per-class.

Methods inherited from class org.opencv.dnn.Model

getNativeObjAddr, predict, setInputCrop, setInputMean, setInputParams, setInputParams, setInputParams, setInputParams, setInputParams, setInputParams, setInputScale, setInputSize, setInputSize, setInputSwapRB, setPreferableBackend, setPreferableTarget

Methods inherited from class java.lang.Object

clone, equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

DetectionModel

protected DetectionModel(long addr)

DetectionModel

public DetectionModel(String model,
                      String config)

Create detection model from network represented in one of the supported formats. An order of model and config arguments does not matter.

Parameters:

model - Binary file contains trained weights.

config - Text file contains network configuration.

DetectionModel

public DetectionModel(String model)

Create detection model from network represented in one of the supported formats. An order of model and config arguments does not matter.

Parameters:

model - Binary file contains trained weights.

DetectionModel

public DetectionModel(Net network)

Create model from deep learning network.

Parameters:

network - Net object.

Method Detail

__fromPtr__

public static DetectionModel __fromPtr__(long addr)

setNmsAcrossClasses

public DetectionModel setNmsAcrossClasses(boolean value)

nmsAcrossClasses defaults to false, such that when non max suppression is used during the detect() function, it will do so per-class. This function allows you to toggle this behaviour.

Parameters:

value - The new value for nmsAcrossClasses

Returns:

automatically generated

getNmsAcrossClasses

public boolean getNmsAcrossClasses()

Getter for nmsAcrossClasses. This variable defaults to false, such that when non max suppression is used during the detect() function, it will do so only per-class

Returns:

automatically generated

detect

public void detect(Mat frame,
                   MatOfInt classIds,
                   MatOfFloat confidences,
                   MatOfRect boxes,
                   float confThreshold,
                   float nmsThreshold)

Given the input frame, create input blob, run net and return result detections.

Parameters:

classIds - Class indexes in result detection.

confidences - A set of corresponding confidences.

boxes - A set of bounding boxes.

confThreshold - A threshold used to filter boxes by confidences.

nmsThreshold - A threshold used in non maximum suppression.

frame - automatically generated

detect

public void detect(Mat frame,
                   MatOfInt classIds,
                   MatOfFloat confidences,
                   MatOfRect boxes,
                   float confThreshold)

Given the input frame, create input blob, run net and return result detections.

Parameters:

classIds - Class indexes in result detection.

confidences - A set of corresponding confidences.

boxes - A set of bounding boxes.

confThreshold - A threshold used to filter boxes by confidences.

frame - automatically generated

detect

public void detect(Mat frame,
                   MatOfInt classIds,
                   MatOfFloat confidences,
                   MatOfRect boxes)

Given the input frame, create input blob, run net and return result detections.

Parameters:

classIds - Class indexes in result detection.

confidences - A set of corresponding confidences.

boxes - A set of bounding boxes.

frame - automatically generated

finalize

protected void finalize()
                 throws Throwable

Overrides:

finalize in class Model

Throws:

Throwable