Abstract: Image segmentation is the fundament of image analysis，understanding and machine vision.To consider man’s vision subjective，image processing is suitable to use fuzzy techniques.In this paper，the criteria function of choosing image segmentation threshold is defined on the base of fuzzy set’s subsethood theory.Experimental results show that image by thresholding segmentation method based on subsethood theory is feasible，and the segmentation performance of it is obviously better than that of thresholding method based on fuzzy entropy or similarity.

Key words: image segmentation, thesholding method, subsethood, fuzzy entropy, friendship function

摘要： 图像分割是图像分析、理解和机器视觉的基础。由于人眼视觉的主观性使得图像比较适合使用模糊技术进行处理。基于模糊集的包含度理论，定义了图像分割选取阈值的准则函数。实验结果表明，提出的图像阈值化分割方法是可行的，且分割性能明显优于基于模糊熵或贴近度的分割法。

关键词: 图像分割, 阈值法, 包含度, 模糊熵, 隶属函数