net.imglib2.algorithm.edge

Class SubpixelEdgelDetection

java.lang.Object

net.imglib2.algorithm.edge.SubpixelEdgelDetection

public class SubpixelEdgelDetection
extends Object

Compute the set of sub-pixel edgels for the input image. A pixel contains an edgel if its gradient magnitude is a local maximum in the direction of the gradient (perpendicular to the edge). The sub-pixel position of the edgel is found as the maximum of a quadratic function fitted to the neighboring magnitudes in the gradient direction. The method is described in

F. Devernay, A Non-Maxima Suppression Method for Edge Detection with Sub-Pixel Accuracy , RR 2724, INRIA, nov. 1995

Author:

Tobias Pietzsch

Constructor Summary

Constructors

Constructor and Description
SubpixelEdgelDetection()

Method Summary

Modifier and Type	Method and Description
static <T extends net.imglib2.type.numeric.RealType<T>> ArrayList<Edgel>	getEdgels(net.imglib2.RandomAccessibleInterval<T> input, net.imglib2.img.ImgFactory<T> factory, double minGradientMagnitude) Compute the set of sub-pixel edgels for the input image.

Methods inherited from class java.lang.Object

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

Constructor Detail

SubpixelEdgelDetection

public SubpixelEdgelDetection()

Method Detail

getEdgels

public static <T extends net.imglib2.type.numeric.RealType<T>> ArrayList<Edgel> getEdgels(net.imglib2.RandomAccessibleInterval<T> input,
                                                                                          net.imglib2.img.ImgFactory<T> factory,
                                                                                          double minGradientMagnitude)

Compute the set of sub-pixel edgels for the input image. A pixel contains an edgel if its gradient magnitude is a local maximum in the direction of the gradient (perpendicular to the edge). The sub-pixel position of the edgel is found as the maximum of a quadratic function fitted to the neighboring magnitudes in the gradient direction.

Note: The input image type must be a signed type! Otherwise gradient computation will not work.

Parameters:

input - input image

factory - used to create an image to store the partial derivatives of the input image

minGradientMagnitude - Only consider a pixel an edgel if the gradient magnitude is greater than this

Returns:

set of edgels found in the input image