Open AccessAdaptive empirical Bayes filter
Author(s) -
Deng G.
Publication year - 2017
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
ISSN - 1350-911X
DOI - 10.1049/el.2017.1308
Subject(s) - adaptive filter , bayes' theorem , kernel adaptive filter , boosting (machine learning) , filter (signal processing) , mathematics , filter design , root raised cosine filter , bayes estimator , bayesian probability , algorithm , computer science , recursive bayesian estimation , artificial intelligence , control theory (sociology) , computer vision , control (management)
A new edge‐aware filter called the empirical Bayes filter (EBF) is presented. It is shown that the bilateral filter (BF), being a special case of the EBF, is an optimal filter in terms of Bayesian linear least square estimation. An adaptive EBF (AEBF), which is an adaptive combination of the BF output and the original image, is developed. Experimental results demonstrated that the AEBF outperforms the boosting algorithm in terms of improving the contrast of the bilateral filtered image.