基于最大模糊熵原理的多阈值图像分割新算法

基于最大模糊熵准则,提出了一种新的多阈值图像分割算法。该算法通过定义一种简单的线性模糊隶属度函数,将图像模糊划分为若干个不同的区域;同时采用自适应遗传算法搜索最大模糊熵准则下图像的一组最佳分割阈值,并对遗传算法中的编码方式及交叉算子、变异算子进行了一些有益的改进,极大地减少了计算量和存储空间,加快了算法的搜索速度。通过仿真实验证明该算法对复杂图像良好的分割效果和较强的实时处理能力。

Multilevel thresholding algorithm for image segmentation based on maximum fuzzy entropy

Based on maximum fuzzy entropy, a new multilevel thresholding algorithm for image segmentation is proposed. It divides the image into different sections fuzzily through defining a simple linear fuzzy neighborhood function, and uses adaptive genetic algorithm to find the best threshold. Improvements are made on encoding method and crossover and mutation operators, which can reduce the computation time and storage space required, and increase the searching speed greatly. Experiments show that good results can be achieved on some complex images, and less time is spent by the algorithm.