A new adaptive filter and quality evaluation index for image restoration

Citation
Ar. Jimenez et al., A new adaptive filter and quality evaluation index for image restoration, PATT RECOG, 34(2), 2001, pp. 457-467
Citations number
20
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
PATTERN RECOGNITION
ISSN journal
00313203 → ACNP
Volume
34
Issue
2
Year of publication
2001
Pages
457 - 467
Database
ISI
SICI code
0031-3203(200102)34:2<457:ANAFAQ>2.0.ZU;2-U
Abstract
This paper presents a new technique for image adaptive restoration called 3 sigma -multiresolution plane fitting (3 sigma -MPF) and the global restora tion index (GRI) index, which evaluates the restoration performance of smoo thing methods. The 3 sigma -MPF filter is based on an adaptive plane fittin g technique using variable-sized windows with different behavior for contin uous regions or discontinuities. The criteria used to select the window siz e and detect discontinuities is the 3 sigma -fidelity test. The GRI, integr ates two independent indices measuring restoration smoothness and fidelity. Using this index, the 3 sigma -MPF filter performance is compared with oth er traditional methods, demonstrating better performance under Gaussian noi se and robustness against outliers. (C) 2000 Pattern Recognition Society. P ublished by Elsevier Science Ltd. All rights reserved.