A new prior shape model for level set segmentation
Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. CIARP 2011. Lecture Notes in Computer Science, vol. 7042. Springer, Berlin, Heidelberg, pp. 125–132.
Abstract
An abstract is not available in the lab archive. View the paper at its source.
This 2011 conference paper introduces a local shape parameter and a Gaussian-weighted shape energy for guiding level set image segmentation. The authors report feasibility experiments on synthetic, optical and MR images.
Summary
The paper presents a prior shape model for level set segmentation. It represents the shape model implicitly with a signed distance function and incorporates a local shape parameter to regulate model fitting. The authors define a shape energy based on this parameter and couple it with a Gaussian kernel to weight the shape model.
The authors describe the Gaussian weighting as allowing the evolving level set to deform around the shape model while smoothing along edges. They report experiments on synthetic, optical and MR images as demonstrating the feasibility of the proposed shape model and energy. The abstract does not specify quantitative comparisons or establish clinical validation.
Cite this work
Poay Hoon Lim; Ulas Bagci; Li Bai (2011). A new prior shape model for level set segmentation. Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. CIARP 2011. Lecture Notes in Computer Science, vol. 7042. Springer, Berlin, Heidelberg, pp. 125–132. https://doi.org/10.1007/978-3-642-25085-9_14
BibTeX