摘要
加权模糊C均值算法(WFCM)是1种有效的图像分割算法,该算法的分类数需要人工确定。为了提高算法的自动化程度,本文提出了先采用基于直方图平滑的峰点检测方法自动确定图像的分类数,再利用WFCM分割图像的算法。大量实验结果显示,此算法能够正确地对图像分类,并且合理地分割图像。
Weighted Fuzzy c-means(WFCM) is a very effective algorithm for image segmentation, but the classifying number must be appointed in advance. In this paper, we propose a new algorithm to increase automaticity. It automatically detects the number of the peaks by smoothing the histogram in order to obtain the classifying number; then segment the images with WFCM. The examples in this paper indicate that the new algorithm can accurately classify and effectively segment the images.
出处
《中国海洋大学学报(自然科学版)》
CAS
CSCD
北大核心
2007年第3期485-488,共4页
Periodical of Ocean University of China
基金
国家博士点基金(20040699015)
西北工业大学研究生创业种子基金(Z2000631)资助
关键词
模糊C均值
图像分割
自动分类
fuzzy c-means
image segmentation
auto-classification