摘要
针对激光图像分割处理的问题,提出了一种基于自适应遗传算法的激光图像分割处理算法.该算法将自适应遗传算法与最大类间方差分割方法相结合,将图像类间方差作为适应度函数,利用交叉概率和变异概率动态调整自适应遗传算法求解最大类间方差的最优阈值.为了衡量该算法的处理效果,分别采用本文算法和最大类间方差图像分割算法对图像进行处理.结果表明,该算法的CI值为0. 417,能够对图像进行有效分割,且分割的准确性和运算速率均优于传统的最大类间方差分割方法,具有较高的实践价值.
Aiming at the problem of laser image segmentation processing,a laser image segmentation processing algorithm based on adaptive genetic algorithm was proposed. The algorithm combined the adaptive genetic algorithm with the maximum inter-class variance segmentation method,and the image inter-class variance was taken as the fitness function. The self-adaptive genetic algorithm was dynamically adjusted with both crossover probability and mutation probability to solve the optimal threshold of maximum inter-class variance. In order to evaluate the processing effect of the proposed algorithm,the proposed algorithm and the maximum inter-class variance image segmentation algorithm were used to deal with the images. The results showthat the CI value of the algorithm is 0. 417 and can effectively segment the image,and the segmentation accuracy and operation speed are better than those of traditional maximum inter-class variance segmentation method. The algorithm has high practical value.
作者
周理
刘琰
ZHOU Li;LIU Yan(School of Information Science and Engineering,Fujian University of Technology,Fuzhou 350118,China;School of Mechanical and Electrical Engineering,Hunan City University,Yiyang 413000,China)
出处
《沈阳工业大学学报》
EI
CAS
北大核心
2019年第2期174-178,共5页
Journal of Shenyang University of Technology
基金
国家自然科学基金资助项目(11302051)
福建工程学院校级科研启动项目(GY-Z13117)
关键词
自适应遗传算法
激光图像
图像分割
最大类间方差
交叉概率
变异概率
适应度函数
阈值
adaptive genetic algorithm
laser image
image segmentation
maximum inter-class variance
crossover probability
mutation probability
fitness function
threshold