关键词: 交互式图像分割, 最小交叉熵, 最大熵, 阈值

Abstract: The segmentation approach on one criterion only does not work well for a lot of images.The thresholding methods based on maximum entropy and minimum cross entropy are used widely today in image segmentation，but they often fail to segment some images，so an interactive image segmentation method based on combining maximum entropy and minimum cross entropy is presented in this paper.Firstly it gets the third thresholding method by combining maximum entropy and minimum cross entropy using a simple algorithm so that the optimal threshold value can not only satisfy the former but also meet the latter，then the optimal segmentation is chosen by users in the three methods.Experimental results show that the proposed method can not only get better segmentation performance and better generalization，but is also more practical.

Key words: interactive image segmentation, Minimum Cross Entropy（MCE）, Maximum Entropy（ME）, threshold