Constructor Detail

• Gauss3

  public Gauss3()

Method Detail

• gauss

  public static <S extends NumericType<S>,T extends NumericType<T>> void gauss(double sigma,
                                                                               RandomAccessible<S> source,
                                                                               RandomAccessibleInterval<T> target)
                                                                        throws IncompatibleTypeException

  Apply Gaussian convolution to source and write the result to target. 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.

  Computation may be multi-threaded, according to the current Parallelization context. (By default, it will use the common ForkJoinPool)

  Type Parameters:

  S - source type

  T - target type

  Parameters:

  sigma - standard deviation of isotropic Gaussian.

  source - source image, must be sufficiently padded (e.g. Views.extendMirrorSingle(RandomAccessibleInterval)) to provide values for the target interval plus a border of half the kernel size.

  target - target image

  Throws:

  IncompatibleTypeException - if source and target type are not compatible (they must be either both RealTypes or the same type).

• gauss

  public static <S extends NumericType<S>,T extends NumericType<T>> void gauss(double[] sigma,
                                                                               RandomAccessible<S> source,
                                                                               RandomAccessibleInterval<T> target)
                                                                        throws IncompatibleTypeException

  Apply Gaussian convolution to source and write the result to target. 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.

  Computation may be multi-threaded, according to the current Parallelization context. (By default, it will use the common ForkJoinPool)

  Type Parameters:

  S - source type

  T - target type

  Parameters:

  sigma - standard deviation in every dimension.

  source - source image, must be sufficiently padded (e.g. Views.extendMirrorSingle(RandomAccessibleInterval)) to provide values for the target interval plus a border of half the kernel size.

  target - target image

  Throws:

  IncompatibleTypeException - if source and target type are not compatible (they must be either both RealTypes or the same type).

• gauss

  @Deprecated
  public static <S extends NumericType<S>,T extends NumericType<T>> void gauss(double[] sigma,
                                                                                           RandomAccessible<S> source,
                                                                                           RandomAccessibleInterval<T> target,
                                                                                           int numThreads)
                                                                                    throws IncompatibleTypeException

  Deprecated. Deprecated. Please use gauss(sigma, source, target) instead. The number of threads used to calculate the Gaussion convolution may by set with the Parallelization context, as show in this example:

   
   Parallelization.runWithNumThreads( numThreads,
      () -> gauss( sigma, source, target )
   );
   
   

  Apply Gaussian convolution to source 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.

  Type Parameters:

  S - source type

  T - target type

  Parameters:

  sigma - standard deviation in every dimension.

  source - source image, must be sufficiently padded (e.g. Views.extendMirrorSingle(RandomAccessibleInterval)) to provide values for the target interval plus a border of half the kernel size.

  target - target image

  numThreads - how many threads to use for the computation.

  Throws:

  IncompatibleTypeException - if source and target type are not compatible (they must be either both RealTypes or the same type).

• gauss

  @Deprecated
  public static <S extends NumericType<S>,T extends NumericType<T>> void gauss(double[] sigma,
                                                                                           RandomAccessible<S> source,
                                                                                           RandomAccessibleInterval<T> target,
                                                                                           ExecutorService service)
                                                                                    throws IncompatibleTypeException

  Deprecated. Deprecated. Please use gauss(sigma, source, target) instead. The ExecutorService used to calculate the Gaussion convolution may by set with the Parallelization context, as show in this example:

   
   Parallelization.runWithExecutor( executorService,
      () -> gauss( sigma, source, target )
   );
   
   

  Apply Gaussian convolution to source 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.

  Type Parameters:

  S - source type

  T - target type

  Parameters:

  sigma - standard deviation in every dimension.

  source - source image, must be sufficiently padded (e.g. Views.extendMirrorSingle(RandomAccessibleInterval)) to provide values for the target interval plus a border of half the kernel size.

  target - target image

  service - service providing threads for multi-threading

  Throws:

  IncompatibleTypeException - if source and target type are not compatible (they must be either both RealTypes or the same type).

• halfkernels

  public static double[][] halfkernels(double[] sigma)

• halfkernelsizes

  public static int[] halfkernelsizes(double[] sigma)

• halfkernel

  public static double[] halfkernel(double sigma,
                                    int size,
                                    boolean normalize)

  Returns a gaussian half kernel with the given sigma and size.

  The edges are smoothed by a second degree polynomial. This improves the first derivative and the fourier spectrum of the gaussian kernel.