一种新的灰度图像阈值分割方法

假设灰度直方图服从正态分布且二个子分布的方差相等 ,基于条件分布的最大相关原则等价于Otsu阈值分割方法。本文在灰度直方图服从泊松分布的假设下 ,运用条件分布的最大相关原则提出了一种新的图象分割方法。对四幅图片进行分割处理的实验结果表明 :在有些场合新方法的分割结果优于Otsu方法

A New Thresholding Method for Grey Image Segmentation

It has been proved that the maximization of the likelihood of the conditional distribution of the conditional distribution of the population mixture model under the assumption of Normal distributions with a common variance is equivalent to Otsu's thresholding. A new thresholding method is proposed in this paper, which is deduced by the maximization of the likelihood of the conditional distribution of the population mixture model under the assumption of Poisson distributions. The comparative experimental results show that the new function performs better than Otsu's method in sine cases.