pzerocross

Locates zero crossing.

Synopsis

pzerocross connexity value [-m mask] [im_in|-] [im_out|-]

Description

pzerocross detects the place where the pixel is a transition between two neighbor pixels. A transition is detected when:

A pixel value is greater than value and at least one of its neighbors is lower than value,
A pixel value is lower than value and at least one of its neighbors is greater than value.

The output image im_out is a unsigned char grayscale image (Img2duc or Img3duc).

This operator is mainly used to detect zero crossing of the laplacian. For example, the value is equal to 127 for unsigned char images and 0 for signed long image.

Parameters

connexity defines the neighbor relationship: 4, 8 in 2D or 6, 26 in 3D. This parameter is ignored for graph.
value belongs to gray levels of the input image im_in.

Inputs

im_in: a grayscale image.

Outputs

im_out: a binary image.

Result

Returns SUCCESS or FAILURE.

Examples

Performs an edge detection using the DOG algorithm (Difference Of Gaussian):

   pexponentialfiltering 0.2 tangram.pan a.pan
   pexponentialfiltering 0.8 tangram.pan b.pan
   psub a.pan b.pan c.pan
   pzerocross 8 0 c.pan out.pan

C++ prototype

Errc PZeroCross( const Img2duc &im_in, Img2duc &im_out, int connexity, Uchar value );

Version française

Localisation des changements de signe des valeurs de pixels.

Author: Régis Clouard