二维Tsallis-Havrda-Charvat熵阈值分割的快速递推算法

Fast Recurring Two-dimensional Tsallis-Havrda-Charvat Entropic Thresholding Algorithms

阈值分割是图像分割中简单有效的方法,应用极为广泛。基于熵的阈值选取方法是其中一类颇受关注的方法,二维Tsallis-Havrda-Charvat熵法分割效果好,但因计算量庞大,难以实用。本文提出了二维Tsallis-Havrda-Charvat熵的阈值分割两种不同的快速递推算法,都可将计算复杂性由O(L^4)减少为O(L^2)。文中给出了二维Tsallis-Havrda-Charvat熵两种快速递推算法的分割结果及运行时间,并与原始算法进行了比较。实验结果表明,这两种递推算法都可以大幅度地提高运算速度,运行时间几乎不到原始算法的0.1%。

Thresholding is a simple and efficient technique for image segmentation in digital image processing. It finds wide applications in various areas. The thresholding algorithm based on entropy is one of the most famous methods. The two-dimensional Tsal- lis-Havrda-Charvat entropic thresholding algorithm has a good performance, but due to its large computation, it is hard to be used in re- ality. In this paper, two fast recurring two-dimensional Tsallis-Havrda-Charvat entropic thresholding algorithms, whose computational complexities are both only O（L^2）, are proposed, while the computational complexity of the original algorithm is O（L^4 ）. Using these two recurring algorithms, the results and processing time of the two-dimensional Tsallis-Havrda-Charvat entropic thresholding algorithm are given. Experimental results show that these two recurring algorithms can both greatly reduce the processing time of images, which is less than 0.1% of the original algorithm.