二维熵阈值法的修改及其快速迭代算法

提出二维熵阈值法的一种修改方法和其快速迭代算法。针对传统二维熵阈值法及其递推算法的高计算复杂性的不足，首先对二维直方图所对应的二元概率分布进行修改并得到一种新的二维熵阈值法。其次假设二维直方图所对应的二元概率分布是连续可微的条件下导出的修改后的二维熵阈值法的快速迭代算法。实验结果表明，文中提出的修改二维熵阈值法及其快速迭代算法是可行的，且快速迭代算法的时间消耗相对其递归算法有很大程度地降低。

Modification of Two-Dimensional Entropic Thresholding Method and Its Fast Iterative Algorithm

A modified method for two-dimensional entropic thresholding method and its fast iterative algorithm are proposed.Aiming at the disadvantage of high computational complexity of the classical two-dimensional entropic thresholding and its recursive algorithm,the two-variables probability distribution of two dimensional histogram is firstly modified and a new two-dimensional entropic thresholding method is obtained.Then,the fast iterative algorithm for the new modified two-dimensional entropic thresholding method is educed on the assumption that the modified two-variables probability distribution of two-dimensional histogram is continuous and differentiable.Experimental results show that the modified two-dimensional entropic thresholding method and its fast iterative algorithm are feasible,and the computational time of the fast iterative algorithm is much less than that of its recursive algorithm to a certain extent.