boofcv.alg.shapes.polygon

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

java.lang.Object

boofcv.alg.shapes.polygon.RefinePolygonLineToImage<T>

All Implemented Interfaces:

RefineBinaryPolygon<T>

public class RefinePolygonLineToImage<T extends boofcv.struct.image.ImageGray>
extends java.lang.Object
implements RefineBinaryPolygon<T>

Improves the fits of a polygon's which is darker or lighter than the background. Polygon's edges are assumed to be perfectly straight lines. The edges are processed individually and fit to a line using weighted regression. Both black squares with white backgrounds and white shapes with black backgrounds can be found. The edges are selected to maximize the difference between light and dark regions.

For example, assume an image axis aligned rectangle has a lower extent of 1,2 and a upper extent of 12,15, is entirely filled, excluding the upper extent pixels (as is typical). Then the found lower and upper extends of the found polygon will also be 1,2 and 12,15.

If a line lies entirely along the image border it is not modified. If part of it lies along the image then only points not near the border are used to optimize its location.

For input polygons which are in undistorted coordinates by with a distorted image call getSnapToEdge() and invoke BaseIntegralEdge.setTransform(PixelTransform2_F32).

Field Summary

Fields

Modifier and Type	Field and Description
protected T	image

Constructor Summary

Constructors

Constructor and Description
RefinePolygonLineToImage(double cornerOffset, int lineSamples, int sampleRadius, int maxIterations, double convergeTolPixels, double maxCornerChangePixel, java.lang.Class<T> imageType) Constructor which provides full access to all parameters.
RefinePolygonLineToImage(int numSides, java.lang.Class<T> imageType) Simplified constructor which uses reasonable default values for most variables

Method Summary

Modifier and Type	Method and Description
void	clearLensDistortion() Clears the previously set lens distortion
SnapToLineEdge<T>	getSnapToEdge()
protected boolean	optimize(georegression.struct.point.Point2D_F64 a, georegression.struct.point.Point2D_F64 b, georegression.struct.line.LineGeneral2D_F64 found) Fits a line defined by the two points.
protected boolean	optimize(georegression.struct.shapes.Polygon2D_F64 seed, georegression.struct.shapes.Polygon2D_F64 current) Refines the initial line estimates using EM.
boolean	refine(georegression.struct.shapes.Polygon2D_F64 input, java.util.List<georegression.struct.point.Point2D_I32> contour, org.ddogleg.struct.GrowQueue_I32 splits, georegression.struct.shapes.Polygon2D_F64 output) Refines the fit a polygon by snapping it to the edges.
void	setImage(T image) Sets the image which is going to be processed.
void	setLensDistortion(int width, int height, boofcv.struct.distort.PixelTransform2_F32 distToUndist, boofcv.struct.distort.PixelTransform2_F32 undistToDist) Specifies lens distortion
void	setTransform(boofcv.struct.distort.PixelTransform2_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

Field Detail

image

protected T extends boofcv.struct.image.ImageGray image

Constructor Detail

RefinePolygonLineToImage

public RefinePolygonLineToImage(double cornerOffset,
                                int lineSamples,
                                int sampleRadius,
                                int maxIterations,
                                double convergeTolPixels,
                                double maxCornerChangePixel,
                                java.lang.Class<T> imageType)

Constructor which provides full access to all parameters. See code documents value a description of these variables.

RefinePolygonLineToImage

public RefinePolygonLineToImage(int numSides,
                                java.lang.Class<T> imageType)

Simplified constructor which uses reasonable default values for most variables

Parameters:

numSides - Number of sides on the polygon

imageType - Type of input image it processes

Method Detail

setImage

public void setImage(T image)

Sets the image which is going to be processed. If a transform is to be used BaseIntegralEdge.setTransform(boofcv.struct.distort.PixelTransform2_F32) should be called before this.

Specified by:

setImage in interface RefineBinaryPolygon<T extends boofcv.struct.image.ImageGray>

setLensDistortion

public void setLensDistortion(int width,
                              int height,
                              boofcv.struct.distort.PixelTransform2_F32 distToUndist,
                              boofcv.struct.distort.PixelTransform2_F32 undistToDist)

Description copied from interface: RefineBinaryPolygon

Specifies lens distortion

Specified by:

setLensDistortion in interface RefineBinaryPolygon<T extends boofcv.struct.image.ImageGray>

clearLensDistortion

public void clearLensDistortion()

Description copied from interface: RefineBinaryPolygon

Clears the previously set lens distortion

Specified by:

clearLensDistortion in interface RefineBinaryPolygon<T extends boofcv.struct.image.ImageGray>

refine

public boolean refine(georegression.struct.shapes.Polygon2D_F64 input,
                      java.util.List<georegression.struct.point.Point2D_I32> contour,
                      org.ddogleg.struct.GrowQueue_I32 splits,
                      georegression.struct.shapes.Polygon2D_F64 output)

Refines the fit a polygon by snapping it to the edges.

Specified by:

refine in interface RefineBinaryPolygon<T extends boofcv.struct.image.ImageGray>

Parameters:

input - (input) Initial estimate for the polygon. CW or CCW ordering doesn't matter.

output - (output) the fitted polygon

contour - List of pixels in the contour

splits - Indexes in the contour where the initial polygon was found

Returns:

true if successful or false if it failed

optimize

protected boolean optimize(georegression.struct.shapes.Polygon2D_F64 seed,
                           georegression.struct.shapes.Polygon2D_F64 current)

Refines the initial line estimates using EM. The number of iterations is fixed.

optimize

protected boolean optimize(georegression.struct.point.Point2D_F64 a,
                           georegression.struct.point.Point2D_F64 b,
                           georegression.struct.line.LineGeneral2D_F64 found)

Fits a line defined by the two points. When fitting the line the weight of the edge is used to determine how influential the point is

Parameters:

a - Corner point in image coordinates.

b - Corner point in image coordinates.

found - (output) Line in image coordinates

Returns:

true if successful or false if it failed

setTransform

public void setTransform(boofcv.struct.distort.PixelTransform2_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..

getSnapToEdge

public SnapToLineEdge<T> getSnapToEdge()