boofcv.alg.fiducial.calib.circle

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

java.lang.Object

boofcv.alg.fiducial.calib.circle.DetectCircleGrid<T>

boofcv.alg.fiducial.calib.circle.DetectCircleRegularGrid<T>

public class DetectCircleRegularGrid<T extends boofcv.struct.image.ImageGray<T>>
extends DetectCircleGrid<T>

Detects regular grids of circles, see below. A valid grid is in counter-clockwise order and if there are multiple possible solution the solution with corner (0,0) closest to the pixel coordinate (0,0) is selected

For each circle there is are four control points. Each control point corresponds to the tangent line connecting the vertical and horizontal neigbors. Tangent points are used since they are invariant under perspective distortion.

Example of a 6 by 4 grid; row, column.

Field Summary

Fields inherited from class boofcv.alg.fiducial.calib.circle.DetectCircleGrid

ellipseDetector, numCols, numRows

Constructor Summary

Constructors

Constructor and Description
DetectCircleRegularGrid(int numRows, int numCols, boofcv.abst.filter.binary.InputToBinary<T> inputToBinary, boofcv.alg.shapes.ellipse.BinaryEllipseDetector<T> ellipseDetector, EllipsesIntoClusters clustering) Creates and configures the detector

Method Summary

Modifier and Type	Method and Description
protected void	configureContourDetector(T gray) Configures the contour detector based on the image size.
protected void	putGridIntoCanonical(EllipseClustersIntoGrid.Grid g) Puts the grid into a canonical orientation
int	totalEllipses(int numRows, int numCols) Computes the number of ellipses on the grid

Methods inherited from class boofcv.alg.fiducial.calib.circle.DetectCircleGrid

getBinary, getCalibrationPoints, getClustering, getClusters, getClustersPruned, getColumns, getEllipseDetector, getGrider, getGrids, getRows, isVerbose, process, setVerbose

Methods inherited from class java.lang.Object

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

Constructor Detail

DetectCircleRegularGrid

public DetectCircleRegularGrid(int numRows,
                               int numCols,
                               boofcv.abst.filter.binary.InputToBinary<T> inputToBinary,
                               boofcv.alg.shapes.ellipse.BinaryEllipseDetector<T> ellipseDetector,
                               EllipsesIntoClusters clustering)

Creates and configures the detector

Parameters:

numRows - number of rows in grid

numCols - number of columns in grid

inputToBinary - Converts the input image into a binary image

ellipseDetector - Detects ellipses inside the image

clustering - Finds clusters of ellipses

Method Detail

configureContourDetector

protected void configureContourDetector(T gray)

Description copied from class: DetectCircleGrid

Configures the contour detector based on the image size. Setting a maximum contour and turning off recording of inner contours and improve speed and reduce the memory foot print significantly.

Specified by:

configureContourDetector in class DetectCircleGrid<T extends boofcv.struct.image.ImageGray<T>>

totalEllipses

public int totalEllipses(int numRows,
                         int numCols)

Description copied from class: DetectCircleGrid

Computes the number of ellipses on the grid

Specified by:

totalEllipses in class DetectCircleGrid<T extends boofcv.struct.image.ImageGray<T>>

putGridIntoCanonical

protected void putGridIntoCanonical(EllipseClustersIntoGrid.Grid g)

Puts the grid into a canonical orientation

Specified by:

putGridIntoCanonical in class DetectCircleGrid<T extends boofcv.struct.image.ImageGray<T>>