Title :
An improved threshold selection method for image segmentation
Author :
Yang, Xue Dong ; Gupta, Vipin
Author_Institution :
Dept. of Comput. Sci., Regina Univ., Sask., Canada
Abstract :
An unsupervised optimal multi-threshold selection scheme for image segmentation is presented. This method is clearly an improvement on two existing methods, namely, Otsu´s (1979) optimal multi-threshold method and Wang´s (1991) threshold hierarchy method. The histogram is divided into different classes using interval tree structure by thresholding the histogram at different scale levels σ of Gaussian convolution. The different histogram dominant modes are then fitted by a Gaussian distribution and the intersection of these Gaussian curves are new threshold points for the image
Keywords :
image segmentation; statistical analysis; stochastic processes; tree data structures; Gaussian convolution; Gaussian curves; Gaussian distribution; histogram; image segmentation; interval tree structure; optimal multithreshold selection; threshold points; threshold selection method; Computer science; Convolution; Histograms; Image segmentation; Optimized production technology; Rain; Tires;
Conference_Titel :
Electrical and Computer Engineering, 1993. Canadian Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-2416-1
DOI :
10.1109/CCECE.1993.332182
