pentropybinarization

Performs binarization on image using maximum entropy criterion.

Synopsis

pentropybinarization [-m mask] [im_in|-] [im_out|-]

Description

pentropybinarization classifies pixels of the input image im_in into 2 clusters. The threshold value is determined as the gray level value s that maximizes the total amount of information provided by the background and the foreground separately. Since information is measured by entropy, the total amount of information for threshold s is:

   TE(s) = Eb(s) + Ef(s) { entropy for background + entropy for foreground }
         = ln[P(s)(1-P(s))] - H(s)/P(s) - H'(s)/(1-P(s))
     where P(s) = SUM{i=0->s} [p(i)]
     and H(s) = SUM{i=0->s} [p(i)*ln(pi)]
     and H'(s) = SUM{i=s->m-1} [(p(i)*ln(pi)]
     and W*H is the number of pixels
     and m is the number of gray levels.
     and pi = fi/W*H

The maximum entropy criterion is to determine the threshold smax such that:

   TE(smax) = max TE(s)

Inputs

im_in: a grayscale image of bytes (Img2duc, Img3duc).

Outputs

im_out: a grayscale image of bytes (Img2duc, Img3duc).

Result

Returns the threshold value.

Examples

Segments the tangram pieces:

   pentropybinarization tangram.pan a.pan

C++ Prototype

Errc PEntropyBinarization( const Img2duc &im_in, Img2duc &im_out );

Version française

Binarisation de l'image par maximisation de l'entropie interclasse.

Reference

J-C Yen, F-J Chang, S. Chang, "A New Criterion for Automatic Multilevel Thresholding", IEEE Trans. on Image Processing, vol. 4, no. 3, pp 370-378, 1995.

Author: Régis Clouard