Article: Interactive image segmentation based on level sets of probabilities
| Title | Interactive image segmentation based on level sets of probabilities |
|---|---|
| Authors | Liu, Y1 Yu, Y2 |
| Keywords | Curvature Distance Transform Image Segmentation Level Set Method Statistical Classification |
| Issue Date | 2012 |
| Publisher | I E E E. The Journal's web site is located at http://www.computer.org/tvcg |
| Citation | Ieee Transactions On Visualization And Computer Graphics, 2012, v. 18 n. 2, p. 202-213 [How to Cite?] DOI: http://dx.doi.org/10.1109/TVCG.2011.77 |
| Abstract | In this paper, we present a robust and accurate algorithm for interactive image segmentation. The level set method is clearly advantageous for image objects with a complex topology and fragmented appearance. Our method integrates discriminative classification models and distance transforms with the level set method to avoid local minima and better snap to true object boundaries. The level set function approximates a transformed version of pixelwise posterior probabilities of being part of a target object. The evolution of its zero level set is driven by three force terms, region force, edge field force, and curvature force. These forces are based on a probabilistic classifier and an unsigned distance transform of salient edges. We further propose a technique that improves the performance of both the probabilistic classifier and the level set method over multiple passes. It makes the final object segmentation less sensitive to user interactions. Experiments and comparisons demonstrate the effectiveness of our method. © 2012 IEEE. |
| ISSN | 1077-2626 2011 Impact Factor: 2.215 2011 SCImago Journal Rankings: 0.073 |
| DOI | http://dx.doi.org/10.1109/TVCG.2011.77 |
| References | References in Scopus |
| dc.contributor.author | Liu, Y | ||||
|---|---|---|---|---|---|
| dc.contributor.author | Yu, Y | ||||
| dc.date.accessioned | 2012-06-26T06:39:35Z | ||||
| dc.date.available | 2012-06-26T06:39:35Z | ||||
| dc.date.issued | 2012 | ||||
| dc.description.abstract | In this paper, we present a robust and accurate algorithm for interactive image segmentation. The level set method is clearly advantageous for image objects with a complex topology and fragmented appearance. Our method integrates discriminative classification models and distance transforms with the level set method to avoid local minima and better snap to true object boundaries. The level set function approximates a transformed version of pixelwise posterior probabilities of being part of a target object. The evolution of its zero level set is driven by three force terms, region force, edge field force, and curvature force. These forces are based on a probabilistic classifier and an unsigned distance transform of salient edges. We further propose a technique that improves the performance of both the probabilistic classifier and the level set method over multiple passes. It makes the final object segmentation less sensitive to user interactions. Experiments and comparisons demonstrate the effectiveness of our method. © 2012 IEEE. | ||||
| dc.description.nature | published_or_final_version | ||||
| dc.identifier.citation | Ieee Transactions On Visualization And Computer Graphics, 2012, v. 18 n. 2, p. 202-213 [How to Cite?] DOI: http://dx.doi.org/10.1109/TVCG.2011.77 | ||||
| dc.identifier.doi | http://dx.doi.org/10.1109/TVCG.2011.77 | ||||
| dc.identifier.epage | 213 | ||||
| dc.identifier.hkuros | 200757 | ||||
| dc.identifier.isi | WOS:000298043100003
Funding Information: We would like to thank the reviewers, whose comments have been valuable in improving our manuscript. This work was partially supported by US National Science Foundation (NSF) (IIS 09-14631). | ||||
| dc.identifier.issn | 1077-2626 2011 Impact Factor: 2.215 2011 SCImago Journal Rankings: 0.073 | ||||
| dc.identifier.issue | 2 | ||||
| dc.identifier.scopus | eid_2-s2.0-83855162158 | ||||
| dc.identifier.spage | 202 | ||||
| dc.identifier.uri | http://hdl.handle.net/10722/152486 | ||||
| dc.identifier.volume | 18 | ||||
| dc.language | eng | ||||
| dc.publisher | I E E E. The Journal's web site is located at http://www.computer.org/tvcg | ||||
| dc.publisher.place | United States | ||||
| dc.relation.ispartof | IEEE Transactions on Visualization and Computer Graphics | ||||
| dc.relation.references | References in Scopus | ||||
| dc.rights | Creative Commons: Attribution 3.0 Hong Kong License | ||||
| dc.subject | Curvature | ||||
| dc.subject | Distance Transform | ||||
| dc.subject | Image Segmentation | ||||
| dc.subject | Level Set Method | ||||
| dc.subject | Statistical Classification | ||||
| dc.title | Interactive image segmentation based on level sets of probabilities | ||||
| dc.type | Article |
Author Affiliations
- University of Electronic Science and Technology of China
- University of Illinois at Urbana-Champaign