摘要
针对机器视觉中的多目标图像分割问题,提出一种适用于多目标物体的图像分割算法.首先对图像进行图像增强预处理;然后采用基于直方图的模糊C均值聚类算法完成分类任务,实现图像的初分割,将分类后的像素作为种子集;最后利用离散正则化的半监督方法得到自动修正分类结果.实验结果表明,与已有的多目标分割算法相比,该算法分割结果更加精确.
To address the problem of multi-objective image segmentation in computer vision, a multi-objectiveimage segmentation method based on fuzzy C-Means and discrete regularization is proposed in this paper.First, the method preprocesses an input image with image enhancement. Secondly, the FCM clusteringalgorithm based on histogram is used to classify the pixels in the images into the different categories andrealize the initial segmentations. Finally a discrete regularization algorithm as a semi-supervised method revisesthe classified results. The experiments demonstrate the superior performance of the proposed methodin terms of segmentation accuracy.
出处
《计算机辅助设计与图形学学报》
EI
CSCD
北大核心
2015年第1期142-146,共5页
Journal of Computer-Aided Design & Computer Graphics
基金
国家教育部博士点基金(20130061110054)
关键词
图像分割
模糊C均值聚类
离散正则化
image segmentation
fuzzy C-means clustering
discrete regularization
