Abstract
Approach to efficient level-set model with shape priors for images segmentation is considered. The use of edge based level-set model in combination with principal component analysis (PCA) based shape priors for image segmentation is investigated. Shape priors considered as a tool to cope with proper segmentation of overlapping or partially visible objects on input image. It is argued that in some cases consequent optimization of different groups of parameters can be advantageous in comparison to simultaneous optimization of all parameters. The approach was applied for segmentation of fractographic images obtained by scanning electron microscope (SEM). Experimental results for image segmentation using level-set model with shape priors are presented.
Publisher
National Academy of Sciences of Ukraine (Co. LTD Ukrinformnauka)
Reference12 articles.
1. 1. Mumford, D.; Shah, J. Optimal approximations by piecewise smooth functions and associated variational problems. Commun.Pure Appl. Math., 1989, 42(5), 577-685.
2. 2. Chan, T.; Vese, L. "Active contours without edges," IEEE Trans.Image. Process. Feb. 2001, 10(2), 266-277.
3. 3. Caselles, V. R. Kimmel, R.; Sapiro, G. Geodesic active contours. Int. J. Comput. Vis. Feb. 1997, 22(1), 61-79.
4. 4. Lankton, S. D.; Nain, A.; Yezzi, Tannenbaum, A. Hybrid geodesic region-based curve evolutions for imagesegmentation. In Proceedings of the Medical Imaging 2007: Physics of Medical Imaging; Hsieh, J., Flynn, M.J.,Eds.; International Society for Optics and Photonics: Bellingham, WA, USA, 2007, 6510, 65104U
5. 5. Leventon, M.; Grimson, W.; Faugeras, O. Statistical Shape Influence in Geodesic Active Contours. In Proc. IEEE CVPR, 2000.