Utilizing information theory and considering image segmentation from a
communication perspective, the image segmentation process is interpre
ted as a data processing step that operates on a gray-scale image and
produces a segmented image. It is shown that the segmented image conta
ins a certain amount of information about the scene, which is defined
as segmented image information (SII), It is proposed that the SII shou
ld be maximized when an image is thresholded, and this is known as the
maximum segmented image information (MSII) thresholding criterion. Th
e MSII thresholding criterion possesses better properties as compared
with the minimum error (MINE) and the uniform error (UNFE) thresholdin
g criteria, Based on the MSII thresholding criterion, an MSII threshol
ding algorithm is proposed for the thresholding of real images. The MS
II thresholding algorithm is evaluated against several well-known thre
sholding algorithms, The good thresholding results of both synthetic a
nd real images confirm the capabilities of the proposed MSII threshold
ing algorithm. (C) 1998 Academic Press.