A region growing method based on fuzzy connectedness

As a requisite of the digital image process, fuzzy object extraction is of great importance. In this paper, a novel region growing method based on fuzzy connectedness is proposed. Firstly, fuzzy connectedness is taken as similarity measure between the seed pixel and current pixel and guides the region-growing process. Then, a combined stop criterion integrating adaptive threshold and usan area is proposed to keep the homogeneity of the region and prevent over-segmentation. Experimental results show that the proposed method performs excellently for fuzzy object segmentation and outperforms the method of literature [4] in spatial accuracy and robustness.