New Tools for Gray Level Histogram Analysis, Applications in Segmentation

This paper summarizes three algorithms used for the analysis of gray level histograms. Two of them are developed generalizing standard techniques and the other is a completely new method. We will study the advantages of each and give examples of real-world use. Gray level histogram analysis (mainly threshold computation) is a known technique that allows easy and fast segmentation of the regions of interest in an image [1]. Many methods have been proposed for this problem [2], but almost all of them focus only on the problem of bimodal histograms. We will deal with multimodal histograms and, besides, we improve the time efficiency of the most widely used method (Otsu’s [3]).