DMCA
Scale-space and edge detection using anisotropic diffusion (1990)
Cached
Download Links
- [www.cs.huji.ac.il]
- [www.cs.berkeley.edu]
- DBLP
Other Repositories/Bibliography
| Venue: | IEEE Transactions on Pattern Analysis and Machine Intelligence |
| Citations: | 1886 - 1 self |
BibTeX
@ARTICLE{Perona90scale-spaceand,
author = {Pietro Perona and Jitendra Malik},
title = {Scale-space and edge detection using anisotropic diffusion},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {1990},
volume = {12},
pages = {629--639}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract-The scale-space technique introduced by Witkin involves generating coarser resolution images by convolving the original image with a Gaussian kernel. This approach has a major drawback: it is difficult to obtain accurately the locations of the “semantically mean-ingful ” edges at coarse scales. In this paper we suggest a new definition of scale-space, and introduce a class of algorithms that realize it using a diffusion process. The diffusion coefficient is chosen to vary spatially in such a way as to encourage intraregion smoothing in preference to interregion smoothing. It is shown that the “no new maxima should be generated at coarse scales ” property of conventional scale space is pre-served. As the region boundaries in our approach remain sharp, we obtain a high quality edge detector which successfully exploits global information. Experimental results are shown on a number of images. The algorithm involves elementary, local operations replicated over the image making parallel hardware implementations feasible. Index Terms-Adaptive filtering, analog VLSI, edge detection, edge enhancement, nonlinear diffusion, nonlinear filtering, parallel algo-
Keyphrases
edge detection anisotropic diffusion analog vlsi region boundary parallel hardware implementation coarse scale original image resolution image parallel algo new maximum global information new definition index terms-adaptive filtering nonlinear filtering major drawback conventional scale space diffusion process mean-ingful edge edge enhancement diffusion coefficient nonlinear diffusion algorithm involves coarse scale property high quality edge detector approach remain local operation abstract-the scale-space technique gaussian kernel experimental result
