Integration of Fuzzy Connectedness and Fuzzy Clustering for Image Segmentation

Image segmentation is one of the mostchallenging and critical problems in image analysis. Traditional segmentation techniques do not completely meet the challenges mostly posed by inherently fuzzy images. Fuzzy connectedness and fuzzy clustering are considered as two well-known techniques for introducing fuzzy concepts to the problem of image segmentation. Segmentation methods which are based on fuzzy connectedness consider spatial relation of image pixels by “hanging togetherness” a notion based on intensity homogeneity but do not inherently utilize feature information of image pixels. On the other hand, the segmentation domain of fuzzy clustering-based methods is the feature space and as such they do not consider spatial relations among image pixels. In this paper, we present a novel segmentation method based on a combination of fuzzy connectedness and fuzzy clustering via a proposed notion called membership connectedness. By membership connectedness, the spatial relation of image pixels is constructed in membership domain. This construction can be applied after any fuzzy segmentation method Experiments were performed using the synthetic images and both simulated and real brain magnetic resonance image (MRI) datasets. The numerical validation of the results demonstrates the strength of the proposed algorithm especially for medical image segmentation purposes.

کلیدواژه ها:

fuzzy clustering ، fuzzy connectedness ، membership connectedness ، magnetic resonance image ، image segmentation