Translator Disclaimer
3 March 2000 Bayesian multiresolution filter design
Author Affiliations +
Proceedings Volume 3961, Nonlinear Image Processing XI; (2000)
Event: Electronic Imaging, 2000, San Jose, CA, United States
This paper discusses a multiresolution approach to Bayesian design of binary filters. The key problem with Bayesian design is that for any window one needs enough observations of a template across the states of nature to estimate its prior distribution, thus introducing severe constraints on single window Bayesian filter designs. By using a multiresolution approach and optimized training methods, we take advantage of prior probability information in designing large-window multiresolution filters. The key point is that we define each filter value at the largest resolution for which we have sufficient prior knowledge to form a prior distribution for the relevant conditional probability, and move to a sub-window when a non-uniform prior is not available. This is repeated until we are able to make a filtering decision at some window size with a known prior for the probability P(Y equals 1x), which is guaranteed for smaller windows. We consider edge noise for our experiments with emphasis on realistically degraded document images.
© (2000) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Vishnu G. Kamat, Edward R. Dougherty, and Junior Barrera "Bayesian multiresolution filter design", Proc. SPIE 3961, Nonlinear Image Processing XI, (3 March 2000);


Multiresolution filter design
Proceedings of SPIE (March 03 2000)
Real-time large-window binary filter design
Proceedings of SPIE (April 27 2001)
Design of binary filters by sequential basis expansion
Proceedings of SPIE (April 27 2001)
Random sets in image processing
Proceedings of SPIE (September 24 1998)

Back to Top