摘要
最大类间方差法是图像分割中一种常用的阈值分割方法,对于单阈值分割具有显著的效果,但是对于多阈值分割,计算复杂度大、耗时较多。本文将粒子群优化算法与最大类间方差法结合,提出了一种新的图像分割方法,该方法利用粒子群优化算法的寻优高效性,并由灰度图像的最大类间方差值作为适应值,搜索最优分割阈值,实现图像的多阈值分割。实验结果显示,新方法大大缩短了寻找最优阈值的时间,降低了运算复杂度,提高了图像分割速度,说明基于粒子群优化算法的图像分割算法是可行的、有效的。
The maximal variance method is one commonly used method in image segmentation field. It has the remarkable effect for single threshold, but for multilevel thresholds, it has the disadvantage of more complexity and time-consuming. To simplify the com- plexity of maximal variance multilevel threshold method for image segmentation, a novel algorithm based on particle swarm optimization was presented in this paper. Taking the advantage of robustness, adaptability and efficiency of particle swarm optimization and taking the maximal variance of gray image as the fitness, the algorithm could obtain the optimal thresholds. The experiments showed that the algorithm presented in this paper could find better solutions with much little complexity and was feasible and effective.
出处
《测绘科学》
CSCD
北大核心
2010年第2期88-89,122,共3页
Science of Surveying and Mapping
关键词
粒子群优化算法
最大类间方差法
多阈值
图像分割
particle swarm optimization
maximal variance
muhilevel threshold
image segmentation
