On the maximization of the crispness of 2D grayscale histogram for image thresholding

Abstract

In this paper, a novel image thresholding approach based on the crispness maximization of the 2D grayscale histogram is proposed. The threshold vector (T,S) is obtained by an exhaustive search algorithm. In this approach, the difference between these two components is guaranteed to be within a relatively small range. This cannot be achieved in fuzzy entropy based methods, such as Abutaleb, Brink's method. Experimental results show that our proposed approach not only performs well and effectively but also is more robust when applied to noisy images.