摘要
提出了一种新的基于模糊集理论的图像直方图阈值分割算法。该算法首先把图像的直方图预分成目标参考区,背景参考区和模糊区。然后对位于模糊区的灰度级,逐一比较其与目标和背景区的相似度,以决定该灰度级的归属,并最终确定灰度阈值。仿真结果表明:此方法不但能不有效地对灰度图像进行分割,且与传统方法相比更具灵活性。
In this paper,a novel method for gray-level image segmentation based on fuzzy set theory has been proposed and verified.Before thresholding,the histogram of the image is divided into three non-overlapping regions:object reference region, background reference region and fuzzy region.The gray levels in the fuzzy region are compared one by one with both object reference region and background reference region,and decisions whether the gray level belongs to object or background can be made according to the resuh.Experiment shows that the proposed method is efficient and more flexible than traditional methods.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第20期75-77,148,共4页
Computer Engineering and Applications
关键词
图像分割
模糊集
成员函数
模糊度影响因子
阈值
image segmentation
fuzzy set
membership function
fuzziness factor
threshold