基于最大方差法和改进遗传算法的图像分割

针对应用标准遗传算法对一幅灰度图像寻找最优阈值时经常陷入局部寻优的问题,提出了一种利用最大方差法和新的改进遗传算法相结合对图像进行分割的方法.以灰度图像的最大方差作为适应度函数,把图像分割问题变成一个优化问题.利用遗传算法的寻优高效性,搜索到能使分割质量达到最优的分割阈值.实验结果表明,采用新的改进遗传算法和最大方差法相结合对图像搜索全局阈值时能收敛至全局最优解,并且大大缩短寻找最优阈值的时间.

IMAGE SEGMENTATION BASED ON THE METHOD OF THE MAXIMAL VARIANCE AND THE IMPROVED GENETIC ALGORITHM

Aiming to solve the problem of local optimum during the search of the optimal threshold of the.image by using normal genetic algorithm, a new image segmentation method based on the maximal variance method and the improved genetic algorithm is presented. The maximal variance of the gray image is taken as the fitness,and the problem of image segmentation is changed into an optimization problem. Taking the advantage of robustness, adaptability and efficiency of the genetic algorithm, this method can obtain the optimal segmentation result according to different images. Experiment shows that using this method to search for the global threshold can converge the optimal value and decrease the searching time of the threshold.