摘要
由于人像照片的复杂性及模糊C均值聚类算法(FCM)存在着计算量大等问题,将一种改进的模糊聚类方法(FFCM)运用于人像图像分割。再根据模糊分类后的人像照片,提出了适合于人像照片背景替换的目标提取方法。实验表明,这种方法能快速还原模糊分类后的人像目标,并使背景部分替换成其他颜色,从而实现人像照片的背景替换。
Because of the complexity of the portrait images and the heavy calculating burden of the Fuzzy C-Means Clustering（FCM）, an improved Fuzzy C-Means Clustering（FFCM） to the portrait background segmentation was applied in this paper. Besides, this paper proposes the method of object selection, it could quickly revert the original portrait object which having been fuzzy segmented, and change the color of the background successfully.
出处
《计算机应用》
CSCD
北大核心
2006年第2期424-426,共3页
journal of Computer Applications
关键词
图像分割
模糊聚类
分层聚类
图像目标提取
image segmentation
fuzzy clustering
hierarchical clustering
object selection
