摘要
提出了一种基于sup-star模糊推理的自动多级图像分割方法。该方法依据直方图(一维或二维)中蕴含的全局统计信息,通过sup-star模糊推理将所有灰度级聚拢归属在直方图局部极大值所代表的类别中,该方法不必已知分割类数,不需设置分割阈值为其优点。仿真和实际图像的实验结果证明了该方法是行之有效的。
An automatic multilevel image segmentation method is presented. Using the well-known sup-star fuzzy reasoning technique (SSFR), the proposed method aggregates all the gray levels of histogram (1D or 2D) into several classes characterized by the local maximum values of the histogram. The proposed method has the merits of determining the number of the segmentation classes automatically, and avoiding calculating thresholds. Emulating and real image segmentation experiments demonstrate that the SSFR is effective.
出处
《中国体视学与图像分析》
2004年第1期1-6,共6页
Chinese Journal of Stereology and Image Analysis
基金
国家自然基金项目(No.60172066)
国家863计划项目(No.2001AA136070)资助
