Thresholding is an important topic for image processing, pattern recog
nition and computer vision. Selecting thresholds is a critical issue f
or many applications. The fuzzy set theory has been successfully appli
ed to many areas, such as control, image processing, pattern recogniti
on, computer vision, medicine, social science, etc. It is generally be
lieved that image processing bears some fuzziness in nature. In this p
aper, we use the concept of fuzzy c-partition and the maximum fuzzy en
tropy principle to select threshold values for gray-level images. We h
ave conducted experiments on many images. The experimental results dem
onstrate that the proposed approach can select the thresholds automati
cally and effectively, and the resulting images can preserve the main
features of the components of the original images very well. (C) 1998
Pattern Recognition Society. Published by Elsevier Science Ltd. All ri
ghts reserved.