Semi-supervised Segmentation Based on Level Set

The segmentation result of Region-based Active Contour (RAC) usually has little relationship with the initial contour which brings difficulties for Human-Computer Interaction. Based on the interactive information between user and system as the prior knowledge, this paper proposed an interactive object segmentation algorithm RIS. The model proposed is based on the active contour without edge (CV) and the prior shape. As the reason of imprecise prior shape user inputs, the weighted area algorithm based on the gradient and laplace was introduced. To overcome the issue of shrink of the object contour, a new area inspiring item was added into the energy function. The experimental results have shown that, without the prior knowledge model based on the shape training set, the proposed algorithm can segment ROI based on the user inputs information.

[1]  D. Mumford,et al.  Optimal approximations by piecewise smooth functions and associated variational problems , 1989 .

[2]  Tony F. Chan,et al.  Level set based shape prior segmentation , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[3]  Tao Zhang,et al.  Tracking objects using density matching and shape priors , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[4]  Tony F. Chan,et al.  Active contours without edges , 2001, IEEE Trans. Image Process..

[5]  Chunming Li,et al.  Level set evolution without re-initialization: a new variational formulation , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[6]  Mads Nielsen,et al.  Computer Vision — ECCV 2002 , 2002, Lecture Notes in Computer Science.

[7]  Daniel Cremers,et al.  Kernel Density Estimation and Intrinsic Alignment for Shape Priors in Level Set Segmentation , 2006, International Journal of Computer Vision.

[8]  Nikos Paragios,et al.  Shape Priors for Level Set Representations , 2002, ECCV.

[9]  Fabrice Heitz,et al.  Affine-invariant geometric shape priors for region-based active contours , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.