Loop-free snakes for highly irregular object shapes
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review
Author(s)
Detail(s)
Original language | English |
---|---|
Pages (from-to) | 579-591 |
Journal / Publication | Pattern Recognition Letters |
Volume | 23 |
Issue number | 5 |
Publication status | Published - Mar 2002 |
Externally published | Yes |
Link(s)
Abstract
Snakes' are an effective approach to image segmentation. However, self-looping is a common problem that can cause segmentation failure of snakes in the recovery of highly irregular object shapes, such as in long tube-like shapes, sharp corners or deep concave/convex shapes. This paper introduces the notion of loop-free snakes that can quickly and effectively remove all self-loops during their evolution, consistently deforming and conforming to complicated shapes of target objects. The proposed snakes are less sensitive to their initial contour condition, are resilient to their inconsistent parameter settings in a certain degree and require low computing cost in terms of both computation time and storage. Experiments are conducted to segment real images with encouraging results. © 2002 Elsevier Science B.V. All rights reserved.
Research Area(s)
- Deformable model, Segmentation, Self-looping suppression, Snakes, Topological adaptability
Citation Format(s)
Loop-free snakes for highly irregular object shapes. / Ji, Lilian; Yan, Hong.
In: Pattern Recognition Letters, Vol. 23, No. 5, 03.2002, p. 579-591.
In: Pattern Recognition Letters, Vol. 23, No. 5, 03.2002, p. 579-591.
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review