一种基于PSO与OSTU的改进絮体分割算法

An improved floc segmentation algorithm based on PSO and OSTU

介绍了图像分割中常用的直方图法、迭代法、经典大津法的阈值选取原理,然后对水处理混凝过程中的絮体图像进行分割对比实验,结合絮体运动特点和水处理实时性的要求,提出了一种基于粒子群优化(PSO)与OSTU的絮体图像分割的改进算法,即先通过灰度拉伸以增强图像灰度对比,再利用PSO算法的全局搜索能力来改善OSTU方法的阈值选取时间,求出分割阈值。实验表明:该算法能实现絮体图像的准确、快速分割,达到实时计算絮体等效粒径和数量的要求。

Introduce threshold value selection principle such as histogram, iterative and classic OSTU method, then use these three methods to divide floes image of water treatment coagulation process, by comparing the effect and combining the floe movement and the characteristics of the real-time demand in water treatment, put forward a kind of improved OSTU image segmentation method based on particle swarm optimization（PSO）. First, using gray stretch to enhance image gray-scale contrast and then utilizing global search ability of PSO algorithm to improve threshold value time of reinforced OSTU method and get segmentation threshold. Experimental results show that the algorithm can realize accurate and fast segmentation of flocs in water treatment, reach requirement of realtime calculation of equivalent size and number of floe.