基于量子粒子群的广义模糊熵阈值法参数选取

Preferences for generalized fuzzy entropy-based thresholding method based on quantum-behavior particle swarm optimization

针对广义模糊熵阈值分割法中参数m的选择问题,提出了一种结合优化算法的自适应参数选取算法。该算法依据一种图像分割质量评价指标建立目标函数,再基于量子粒子群优化搜索算法在参数的变化空间自适应地搜索最佳参数,同时依据模糊熵最大准则对S型隶属度函数中的三个参数(a,b,d)进行了全局组合寻优,从而建立了一个嵌套的优化搜索过程,实现了广义模糊熵图像阈值分割方法的自动阈值选取。实验表明,该方法对光照不均匀图像有更好的分割效果。

An adaptive preferences method is proposed to find the tropy with optimization algorithm. The novel method builds an objecti parameter rn in the generalized fuzzy enve function based on an image segmentation quality evaluation criterion for searching the optimal parameter of the generalized fuzzy entropy and builds the other objective function based on the maximum fuzzy entropy criterion for searching the membership function parameters （a, b, d） in their variation space with the quantum-behavior particle swarm optimization （QPSO） algorithm,respectively. Therefore,it is a nest searching process and realizes the aim of searching the threshold in generalized fuzzy entropy-based image segmentation method automatically. Experiment results show that the new method can obtain good segmentation results on images with illumination noises.