Title :
ICM method for multi-level thresholding using maximum entropy criterion
Author :
Luo, Xiping ; Tian, Jie
Author_Institution :
AILAB, Acad. Sinica, Beijing, China
Abstract :
The maximum entropy approach is one of the most important threshold selection methods. Many authors avoid the computationally prohibitive problem when the maximum entropy criterion is applied to multi-level threshold selection. This paper deals with this problem using the iterated conditional modes (ICM) method. Experiments show ICM can get satisfactory if not optimal results using our initialization algorithm and solve the problem of computation excellently. Also in this paper, we have modified the fuzzy maximum entropy criterion of Cheng et al., (1998). In addition, we compared our ICM method with the simulated annealing algorithm proposed by them
Keywords :
fuzzy logic; image processing; iterative methods; maximum entropy methods; ICM method; fuzzy maximum entropy criterion; image processing; initialization algorithm; iterated conditional modes; maximum entropy criterion; multi-level thresholding; threshold selection; Annealing; Automation; Computational modeling; Convergence; Electrical capacitance tomography; Entropy; Histograms; Image processing; Search methods; State-space methods;
Conference_Titel :
Image Analysis and Processing, 1999. Proceedings. International Conference on
Conference_Location :
Venice
Print_ISBN :
0-7695-0040-4
DOI :
10.1109/ICIAP.1999.797579