Two-Dimensional Arimoto Entropy Image Thresholding based on Ellipsoid Region Search Strategy
Yaoyong Liu, Shuguang Li · 2010
In this paper, a novel image thresholding based on Two-Dimensional Arimoto Entropy (TDAE), called Ellipsoid Region Search Strategy (ERSS), is proposed. The basic idea is to search the optimal threshold from the reference point obtained by one-dimensional OTSU algorithm. Different from existing approaches, our approach provides a scheme to reduce the number of samples through the existing relatively easy and efficient algorithm. Additionally, this approach improves the performance of image segmentation. The experimental results on image segmentation demonstrate that our method is not only significantly more efficient but also accelerates the computation speed of two-dimensional Arimoto entropic thresholding algorithms.