Two-Dimensional Tsallis Symmetric Cross Entropy Image Threshold Segmentation
Yu Li Jun, Jun Ping Zhou, Yin Xuefeng · 2012
This paper proposed a new two-dimensional Tsallis symmetric cross entropy image threshold Segmentation method. First, the two-dimensional Tsallis symmetric cross entropy is given, and then a fast recursive algorithm is used to search the optimal threshold vector. The algorithms do not ignore the pixel points which fall on the region away from the diagonal in the histogram. So it can obtain a better result when it segments an actual image especially for the image which has more edge points and noise points. In addition, this recursive algorithm reduces the computational complexity, greatly improved the efficient. The experimental results show that, the method in this paper has better performance on both effect and speed.