基于蒙特卡罗方法和粗糙熵标准的图像分割方法

提出了一种新的图像分割方法。这种分割方法首先利用粗糙集理论将图像按照一定的规则划分为大小相等的若干图像子块，而后利用蒙特卡罗方法基本原理对划分的图像子块进行一定规模的随机抽样，以随机抽样所得的图像子块为样本进行粗糙熵计算，用所得最大粗糙熵所对应的灰度值为分割阀值对图像进行分割，在采用较小的图像子块划分以取得更好的分割效果的情况下，极大的提高了算法的分割速度。通过对测试图像的MATLAB仿真试验验证了算法在降低计算机消耗方面的有效性，且所得的分割阀值也令人满意。

Image Segmentation Algorithm Based on Monte Carlo Methods and Rough Entropy Standard

The image segmentation algorithm was presented. First of all, it divides the image into some equal size image sub --pieces according with some rules, carries out certain scale random sampling on the image sub--pieces using Monte Carlo Methods, and computes the rough entropy of the sample, and then divides the image by the image gray scale corresponding to the maximum rough entropy. The speed of the image segmentation is largely improved, while adopting the less image sub- pieces to get better segmentation effect. And it has been confirmed in reducing the consumption of the computer is very effective in experiments using MATLAB, and the threshold is satisfied.