pexponentialfiltering

Performs exponential filtering on image.

Synopsis

pexponentialfiltering alpha [-m mask] [im_in|-] [im_out|-]

Description

pexponentialfiltering applies an exponential filter to the input image im_in. The exponential filter is built as follows for one image row, first from left to right (h1) and then from right to left (h2):

   h1[x]= alpha*(im_in[y][x]-h1[x-1]) + h1[x-1]
   h2[x]= alpha*(h1[x]-h2[x+1]) + h[x+1]

The filter is then applied on each rows and each columns.

Parameters

alpha is a real value from the interval [0..1]. It specifies the strength of the filtering:
- 1: light filtering.
- 0: strong filtering.

Inputs

im_in: an image.

Outputs

im_out: an image of the same type as the input 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 PExponentialFiltering( const Img2duc &im_in, Img2duc &im_out, float alpha );

Version française

Lissage par une exponentielle symétrique.

Author: Régis Clouard