Image segmentation by automatic histogram thresholding
Ruey-Ming Chao, Hsien-Chu Wu, Zichun Chen · 2009
Thresholding is an important technology for image segmentation. Before we get the segmentation thresholds, most segmentation technologies need to set many parameters. This paper presents a method to automatically determine how many thresholds should be set and what the best range of each threshold is for different images. It finds the segmentation threshold by observing the change of the variance values and the mean values of each threshold range in the image histogram. The proposed method needs simple calculation, so that it has less time complexity.