CAJ | 학술논문

灰度图像分割问题一般采用传统的最大类间方差法来解决,但是类间方差方法计算量大,不适合实时图像处理.为了解决上述问题.提出了一种改进型遗传退火的阈值分割算法.算法的整个运行过程由冷却温度进度表控制,使用经典的最大类间方差法作为遗传算法的适应度函数,再根据Metropolis准则判断产生的新解是否被接受,从而求得灰度图像的一个最佳阈值.图像分割的仿真结果表明,与传统的最大类间方差法相比,计算量不大,算法具有很强的全局优化搜索能力,由于算法效率高,收敛速度快,适用于实时性的灰度图像处理.
회도도상분할문제일반채용전통적최대류간방차법래해결,단시류간방차방법계산량대,불괄합실시도상처리.위료해결상술문제.제출료일충개진형유전퇴화적역치분할산법.산법적정개운행과정유냉각온도진도표공제,사용경전적최대류간방차법작위유전산법적괄응도함수,재근거Metropolis준칙판단산생적신해시부피접수,종이구득회도도상적일개최가역치.도상분할적방진결과표명,여전통적최대류간방차법상비,계산량불대,산법구유흔강적전국우화수색능력,유우산법효솔고,수렴속도쾌,괄용우실시성적회도도상처리.
In general the traditional Otsu method is used to solve the gray image division problem,but this method is not suitable for real-time image processing because of large computation.In order to solve this problem,thispaper mainly proposes a threshold selection algorithm based on the GASA method.The whole running process of this algorithm was controlled by the temperature cooling schedule,with the classical Otsu method being used as the fitness function of the genetic algorithm.The Metropolis principle is used to determine whether the new solution is available.After several rounds of computing,an optimal threshold value Was obtained.The image simulation result indicates that this algorithm has stronger optimal searching ability and provides hisher efficiency and faster convergence speed,which makes it an appropriate solution to implement the real-time gray image processing.