thresMlComCur

[ XITE Reference Manual | XITE home ]

Name

thresMlComCur - multi-level thresholding by Reddi using complexity curve

Syntax

 thresMlComCur [-mean] [-med] [-scale <scale>]
               [-nt <nt>] <inimage> [<outimage>]

Description

thresMlComCur performs multi-level thresholding according to the method of Reddi et al., here applying complexity curve instead of histogram (see ref).

Options

-mean
Output is mean of pixels between two thresholds.

-med
Output is median of pixels between two thresholds.

-scale
Output is threshold-level * scale. EX: pixels between 3rd and 4th threshold get value 3*scale. If scale is too large, it is adjusted down to make 255 the maximum output value. Default scale is too large.

-nt no_of_ths
Desired number of thresholds.

Use onlu one of -mean, -med and -scale.

See also

thresMl(1), thresMlAppScale(3), thresMlAppMean(3), thresMlAppMedian(3), thresMlApply(3), histoCentroid(3), thresMlCentroid(3), thresMlCentroid(3), mkComCur(3), thresMlComCur(3), thresMlReddi(3), thresMlReddi(1), thresMlWaHa(3), thresMlWaHa(1)

Restrictions

inimage must have bands with pixel type unsigned byte. 1 <= thresholds <= 255.

Reference

S. S. Reddi, S. F. Rudin and H. R. Keshavan
"An Optimal Multiple Threshold Scheme for Image Segmentation" IEEE Transactions on Systems, Man and Cybernetics, Vol SMC-14, pp 661-665, 1984.

Sei-ichiro Kamata, Richard O. Eason and Eiji Kawaguchi
"Complexity Curves Versus Histograms and Their Application to Image Segmentation", 1070-1077

Author

Hung Buu Huynh, BLAB, Ifi, UiO

Examples

 # Find 4 thersholds values of mona.img
 thresMlComCur  -nt 4  mona.img

Id

$Id: thresMlComCur.c,v 1.17 1997/01/14 13:07:08 svein Exp $