摘要
遗传算法是一种软计算方法 ,它作为一种新的全局优化搜索算法用于图象处理 ,是一种具有发展潜力的智能信息处理方法 .对于利用遗传算法处理灰度图象来说 ,由于其数据量和计算量很大 ,并且传统遗传算法存在“过早收敛”的问题 ,使得图象恢复质量不理想 .该文针对医学内窥镜摄取的图象纹理和病灶特征 ,设计了一种新的二维染色体编码方法 ,并将模拟退火和遗传算法相结合的算法引入灰度图象恢复过程中的遗传操作 ,实验结果表明 ,该方法对解决由“过早收敛”所造成的图象恢复质量不高的问题 。
Genetic algorithm is a soft computing method and new optimization searching algorithm. Being used in processing image, it is an intellective information processing method. When processing grayscale image, because the amount of data and computing of image is very large and traditional genetic algorithm has early convergence problems, the quality of image restoration is wrong. The paper designs a new chromsome coding method, and introduces the combination of simulated annealing and genetic algorithm into genetic operation in restoring of grayscale image. This new method of image restoration by the paper presented solved the problem that the quality of image restoration is bad, because of early convergence in traditional genetic algorithm. By experimenting of medicine image, it's results show the quality of image restoration using this new method is good than traditional genetic algorithm.
出处
《中国图象图形学报（A辑）》
CSCD
2000年第4期327-330,共4页
Journal of Image and Graphics
关键词
遗传算法
图象处理
图象恢复
医学影像诊断
Genetic algorithm, Image processing, Image restoration, Simulated annealing algorithm
