Directional gradient vector flow for snakes

• Main Authors: Krishnan, Shankar M., Cheng, Jierong, Foo, Say Wei
• Other Authors: IEEE International Symposium on Signal Processing and Information Technology (4th : 2004 : Rome, Italy)
• Format: Conference or Workshop Item
• Language: English
• Published: 2009
• Online Access: https://hdl.handle.net/10356/91602; http://hdl.handle.net/10220/4629
• Institution: Nanyang Technological University

Snakes, or active contour models, have been widely used in image segmentation. However, most present snake models cannot discern between positive and negative step edges. In this paper, a new type of dynamic external force for snakes named directional gradient vector flow (DGVF) is proposed to solve this problem by incorporating directional gradient information. It makes use of the gradients in both x and y directions and deals with the external force field for the two directions separately. In snake deformation, the DGVF field is utilized dynamically according to the orientation of snake in each iteration. Experiment results demonstrate that the DGVF snake provides a better segmentation than GVF snake in situations when edges of different directions are present and may pose confusion for segmentation.