Image Segmentation Using Curve Evolution and Anisotropic Diffusion: An Integrated Approach

In this paper, a new model is proposed for image segmentation that integrates the curve evolution and anisotropic diffusion methods. The curve evolution method, utilizing both gradient and region information, segments an image into multiple regions. During the evolution of the curve, anisotropic diffusion is adaptively applied to the image to remove noise while preserving boundary information. Coupled partial differential equations (PDE’s) are used to implement the method. Experimental results show that the proposed model is successful for complex images with high noise. Key words: Image segmentation, curve evolution, anisotropic diffusion, coupled PDEs. 

• Type: Conference
• Year: 2005
• Where: ISM
• Authors: Yongsheng Pan, J. Douglas Birdwell, Seddik M. Djouadi
• Keywords: Anisotropic Diffusion, Curve Evolution, Image, ISM 2005