boofcv.alg.shapes.edge

Class PolygonEdgeIntensity<T extends boofcv.struct.image.ImageGray>

java.lang.Object

boofcv.alg.shapes.edge.PolygonEdgeIntensity<T>

public class PolygonEdgeIntensity<T extends boofcv.struct.image.ImageGray>
extends java.lang.Object

Looks at the difference in pixel values along the edge of a polygon and decides if its a false positive or not. The average difference along the polygons edge is the score. Note that the abs is only taken after the sum is finished, so objects which are entirely dark/light along the edge will have an advantage.

Constructor Summary

Constructors

Constructor and Description
PolygonEdgeIntensity(double cornerOffset, double tangentDistance, int numSamples, java.lang.Class<T> imageType) Constructor which configures scoring.

Method Summary

Modifier and Type	Method and Description
boolean	checkIntensity(boolean insideDark, double threshold) Checks the edge intensity against a threshold.
boolean	computeEdge(georegression.struct.shapes.Polygon2D_F64 polygon, boolean ccw) Checks to see if its a valid polygon or a false positive by looking at edge intensity
double	getAverageInside()
double	getAverageOutside()
double	getCornerOffset()
double	getTangentDistance()
void	setCornerOffset(double cornerOffset)
void	setImage(T image) Sets the image which is going to be processed.
void	setTangentDistance(double tangentDistance)
void	setTransform(boofcv.struct.distort.PixelTransform_F32 undistToDist) Used to specify a transform that is applied to pixel coordinates to bring them back into original input image coordinates.

Methods inherited from class java.lang.Object

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

Constructor Detail

PolygonEdgeIntensity

public PolygonEdgeIntensity(double cornerOffset,
                            double tangentDistance,
                            int numSamples,
                            java.lang.Class<T> imageType)

Constructor which configures scoring.

Parameters:

cornerOffset - Number of pixels away from corner it will start sampling

tangentDistance - How far from the line it will sample tangentially

numSamples - Number of points it will sample along an edge

imageType - Type of image it will process

Method Detail

setTransform

public void setTransform(boofcv.struct.distort.PixelTransform_F32 undistToDist)

Used to specify a transform that is applied to pixel coordinates to bring them back into original input image coordinates. For example if the input image has lens distortion but the edge were found in undistorted coordinates this code needs to know how to go from undistorted back into distorted image coordinates in order to read the pixel's value.

Parameters:

undistToDist - Pixel transformation from undistorted pixels into the actual distorted input image..

setImage

public void setImage(T image)

Sets the image which is going to be processed.

computeEdge

public boolean computeEdge(georegression.struct.shapes.Polygon2D_F64 polygon,
                           boolean ccw)

Checks to see if its a valid polygon or a false positive by looking at edge intensity

Parameters:

polygon - The polygon being tested

ccw - True if the polygon is counter clockwise

Returns:

true if it could compute the edge intensity, otherwise false

checkIntensity

public boolean checkIntensity(boolean insideDark,
                              double threshold)

Checks the edge intensity against a threshold. dark: outside-inside ≥ threshold light: inside-outside ≥ threshold

Parameters:

insideDark - is the inside of the polygon supposed to be dark or light?

threshold - threshold for average difference

Returns:

true if the edge intensity is significant enough

getCornerOffset

public double getCornerOffset()

setCornerOffset

public void setCornerOffset(double cornerOffset)

getTangentDistance

public double getTangentDistance()

setTangentDistance

public void setTangentDistance(double tangentDistance)

getAverageInside

public double getAverageInside()

getAverageOutside

public double getAverageOutside()