作者: Chengxin Yan , Nong Sang , Tianxu Zhang
DOI: 10.1016/S0167-8655(03)00154-5
关键词:
摘要: Transition region based thresholding is a newly developed approach for image segmentation in recent years. Gradient-based transition extraction methods (G-TREM) are greatly affected by noise. Local entropy information theory represents the variance of local and catches natural properties regions. In this paper, we present novel entropy-based method (LE-TREM), which effectively reduces affects Experimental results demonstrate that LE-TREM significantly outperforms conventional G-TREM.