A Novel Snake Model for X-Ray Image Segmentation
Chengsu Ouyang, Yongxuan Huang, Jun Yuan · 2009
In image segmentation and computer vision, gradient vector flow (GVF) snake model is used widely. GVF snake has larger capture range and stronger convergence ability to boundary concavities than traditional snake. However the dots outside force field of the GVF field can't converge to the actual objects, and GVF snake becomes sensitive to its initial contour condition. Thus, there are problems in convergence processing to boundaries of the irregular object with highly concavities in human body image. In this paper, a new snake model is proposed, which combines the GVF snake model and attractable snake model. A new self-feedback loop is proposed and a disturbance variable is presented for shrinking the dots outside the rang of force field . The proposed snake model is used for segmenting an X-ray finger image and can provide a satisfied result. In addition, comparisons between the proposed snake and GVF snake show that our model overwhelms GVF snake model.