Convexity Shape Prior for Level Set based Image Segmentation Method
This work addresses the need for convex segmentation in medical or object recognition tasks, but it is incremental as it builds on existing level set models.
The paper tackles the problem of ensuring convex segmented regions in level set image segmentation by introducing a geometric convexity shape prior, which is enforced as an inequality constraint on the signed distance function. Experimental results demonstrate the effectiveness and quality of the proposed method.
We propose a geometric convexity shape prior preservation method for variational level set based image segmentation methods. Our method is built upon the fact that the level set of a convex signed distanced function must be convex. This property enables us to transfer a complicated geometrical convexity prior into a simple inequality constraint on the function. An active set based Gauss-Seidel iteration is used to handle this constrained minimization problem to get an efficient algorithm. We apply our method to region and edge based level set segmentation models including Chan-Vese (CV) model with guarantee that the segmented region will be convex. Experimental results show the effectiveness and quality of the proposed model and algorithm.