Edge-based image segmentation with MOVels
- 20 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
We present a new technique for edge based segmentation, which is fast and accurate even in the presence of nested contours. The shape characteristics of the contours are perfectly recovered, with no smoothing, and only closed contours are returned; this leads to a well-defined segmentation of the images. The algorithm is easily parallelizable, and is well-suitable for specialized, real time hardware. Experimental results are presented, in comparison with the Canny algorithm (J.F. Canny, 1986).Keywords
This publication has 6 references indexed in Scilit:
- Fast and accurate edge-based segmentation with no contour smoothing in 2-D real imagesIEEE Transactions on Image Processing, 2000
- A unified approach to noise removal, image enhancement, and shape recoveryIEEE Transactions on Image Processing, 1996
- Periodic quasi-orthogonal spline bases and applications to least-squares curve fitting of digital imagesIEEE Transactions on Image Processing, 1996
- Shape modeling with front propagation: a level set approachPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1995
- Deformable contours: modeling and extractionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1994
- A Computational Approach to Edge DetectionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1986