A novel local energy-based with fuzzy clustering active contours for image segmentation

作者：Hu Wenjin; Li Zhanming^*; Liu Zhongmin

来源：Journal of Computational Information Systems, 2012, 8(7): 2991-2998.

• Active contours
• Energy functions
• Euler-Lagrange equations
• Fast algorithms
• Fuzzy energy
• Fuzzy information
• Inhomogeneous images
• Intensity inhomogeneity
• Local energy
• Local minimums
• Object boundaries
• Region-based models
• Region-scalable fitting energy

This paper presents a novel local energy function method which integrates fuzzy information of image for image segmentation. The method establishes a new local fuzzy energy function model by using the subordinate property of the membership function in fuzzy clustering and the local intensity statistics to guide the motion of the contours; thus allows the model to deal with intensity inhomogeneity. In order to deal with the problem that local region-based model got stuck in local minimums, intensity re-weighting was adopted. Furthermore, instead of solving the Euler-Lagrange equation, the paper directly calculates the alterations of the fuzzy energy. By this way, the contour converges quickly to the object boundary. Experimental results on images validate the effectiveness of the model when working with intensity inhomogeneous images.

• 出版日期2012