psusan

Performs Susan corner detection.

Synopsis

psusan threshold [-m mask] [im_in|-] [im_out|-]

Description

psusan is a corner detector. Corners are T, L or Y junctions or points with strong texture variation. Corners correspond to double discontinuities of the intensity function caused by discontinuities in the reflectance or the depth functions.

The susan algorithm consists in:

Center a circular mask of radius 3 on each pixel.
Compute the number of pixels in the mask with the same value than the center.
Threshold to produce the image with the strength of each pixel.
Suppress all non maximum pixels to keep only corners.

The output image im_out is a Long image that encodes for each pixel the strength of the response at this point.

Parameters

threshold determines the maximum contrast between two pixels in the same region. The higher is the threshold, the less there are corners in the result. A typical value is 20.

Inputs

im_in: a 2D image.

Outputs

im_out: a Long image.

Result

Returns SUCCESS or FAILURE.

Examples

Extracts corners from image tangram.pan and superimposes corners on the initial image.

   psusan 20 tangram.pan a.pan
   pbinarization 1000 1e30 a.pan b.pan
   padd b.pan tangram.pan out.pan

C++ prototype

Errc PSusan( const Img2duc &im_in, Img2dsl &im_out, int threshold );

Version française

Détection de points d'intérêt selon l'algorithme SUSAN.

Author: Régis Clouard