Adaptive nonlocal filtering: A fast alternative to anisotropic diffusion for image enhancement

Authors
Fischl, B Schwartz, EL
Citation
B. Fischl et El. Schwartz, Adaptive nonlocal filtering: A fast alternative to anisotropic diffusion for image enhancement, IEEE PATT A, 21(1), 1999, pp. 42-48
Citations number
35
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
ISSN journal
01628828 → ACNP
Volume
21
Issue
1
Year of publication
1999
Pages
42 - 48
Database
ISI
SICI code
0162-8828(199901)21:1<42:ANFAFA>2.0.ZU;2-3
Abstract
Nonlinear anisotropic diffusion algorithms provide significant improvement in image enhancement as compared to linear filters. However, the excessive computational cost of solving nonlinear PDEs precludes their use in real-ti me vision applications. In the present paper, we show that two orders of ma gnitude speed improvement is provided by a new image filtering paradigm in which an adaptively determined vector field specifies nonlocal application points for an image filter.