A Computational Approach to Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Handbook of discrete and computational geometry
Handbook of discrete and computational geometry
Normalized Cuts and Image Segmentation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Mean Shift: A Robust Approach Toward Feature Space Analysis
IEEE Transactions on Pattern Analysis and Machine Intelligence
Spectral Segmentation with Multiscale Graph Decomposition
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
Vectorized image segmentation via trixel agglomeration
Pattern Recognition
Rectification of the chordal axis transform skeleton and criteria for shape decomposition
Image and Vision Computing
| Hi-index | 0.00 |
Image contours, obtained by edge detection algorithms, provide a sparse but informative structural representation of image content. However, contours from edge detectors are typically incomplete. Delaunay triangulation of image contour points establishes proximity-based regional bindings of contour elements and supports perceptually meaningful completions of contours for image segmentation. Further, constrained Delaunay triangulations of discretely sampled closed contours can be used to characterize shapes in terms of their parts. This approach to feature extraction, using only contour pixels and their triangulations, offers significant data reduction and computational efficiency for rapid image understanding tasks. In this paper we present a framework for perceptual image analysis that uses proximity properties of Delaunay triangulation as a natural grid for both image segmentation and shape analysis.