摘要
提出基于微分进化和加权模糊C均值的图像分割方法.鉴于加权模糊C均值算法对初始值敏感、容易陷入局部最优的缺点,采用微分进化进行全局寻优,能够找到全局最优的模糊划分;考虑到欧氏距离的局限性,实验中引入特征距离,改善了图像分割效果.实验结果表明,该方法的收敛速度更快、稳定性更好,所获得的适应度值更高.
With superiorities of differential evolution（DE）and weighted fuzzy C means（WFCM）,a new image segmentation method is proposed-image segmentation method based on DE and WFCM.As WFCM is sensitive to initial values and easy to fall into a local optimum,DE is applied for global optimization,so the best globally fuzzy segmentation can be found.Considering the limitation of Euclidian distance,feature distance is applied in the experiment to improve image segmentation.Experiments show that the method has better effect and stability,and reach the better fitness.
出处
《福州大学学报（自然科学版）》
CAS
CSCD
北大核心
2010年第1期36-40,共5页
Journal of Fuzhou University(Natural Science Edition)
基金
福建省自然科学基金资助项目(2009J01283
2009J01248)
福建省科技计划重点资助项目(2008H0026)
关键词
微分进化
加权模糊C均值
特征距离
图像
分割
differential evolution
weighted fuzzy C means
feature distance
image
segmentation
