net.imglib2.algorithm.convolution.kernel

Class SeparableKernelConvolution

java.lang.Object

net.imglib2.algorithm.convolution.kernel.SeparableKernelConvolution

public class SeparableKernelConvolution
extends Object

Convolution with a customizable separable kernel.

Author:

Matthias Arzt

Constructor Summary

Constructors

Constructor and Description
SeparableKernelConvolution()

Method Summary

Modifier and Type	Method and Description
static Convolution<NumericType<?>>	convolution(Kernel1D... kernels) Return an object, that performs the separable convolution with the given kernel.
static Convolution<NumericType<?>>	convolution1d(Kernel1D kernel, int direction) Apply a convolution only in one dimension.
static void	convolve(Kernel1D[] kernels, RandomAccessible<? extends NumericType<?>> source, RandomAccessibleInterval<? extends NumericType<?>> target) Convolve source with a separable kernel and write the result to output.

Methods inherited from class java.lang.Object

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

Constructor Detail

SeparableKernelConvolution

public SeparableKernelConvolution()

Method Detail

convolution

public static Convolution<NumericType<?>> convolution(Kernel1D... kernels)

Return an object, that performs the separable convolution with the given kernel. It additionally allows to set the ExecutorService to use or to query for the required input image size, ...

A small example how it works:

 
 double[][] values = { { 1, 2, 1 }, { -1, 0, 1 } }; int[] center = { 1, 1 };
 Kernel1D[] sobelKernel = Kernel1D.asymmetric( values, center );
 SeparableKernelConvolution.convolution( sobelKernel ).process( Views.extendBorder( inputImage ), outputImage );
 
 

See Also:

Convolution

convolution1d

public static Convolution<NumericType<?>> convolution1d(Kernel1D kernel,
                                                        int direction)

Apply a convolution only in one dimension. For example calculate central differences:

 
 double[] values = { -0.5, 0, 0.5 }; int center = 1;
 Kernel1D[] sobelKernel = Kernel1D.asymmetric( values, center );
 SeparableKernelConvolution.convolution1d( sobelKernel, 0 ).process( Views.extendBorder( inputImage ), outputImage );
 
 

See Also:

Convolution

convolve

public static void convolve(Kernel1D[] kernels,
                            RandomAccessible<? extends NumericType<?>> source,
                            RandomAccessibleInterval<? extends NumericType<?>> target)

Convolve source with a separable kernel and write the result to output. In-place operation (source==target) is supported.

If the target type T is DoubleType, all calculations are done in double precision. For all other target RealTypes float precision is used. General NumericTypes are computed in their own precision. The source type S and target type T are either both RealTypes or both the same type.

Parameters:

kernels - an array containing kernels for every dimension.

source - source image, must be sufficiently padded (use e.g. Views.extendMirrorSingle(RandomAccessibleInterval)) the required source interval.

target - target image.