A hybrid image restoration approach: fuzzy logic and directional weighted median based uniform impulse noise removal |
| |
Authors: | Ayyaz Hussain M Arfan Jaffar Anwar M Mirza |
| |
Affiliation: | 1. Department of Computer Science, FAST National University of Computer and Emerging Sciences, A.K. Brohi Road, H-11/4, Islamabad, Pakistan
|
| |
Abstract: | In this paper, a hybrid image restoration technique based on fuzzy logic and directional weighted median is presented. The
proposed technique consists of noise detection and fuzzy filtering processes to detect and remove uniform (random-valued)
impulse noise while preserving the image details efficiently. In order to preserve image details such as edges and texture
information, a two-stage robust noise detection is presented in this paper. Pixels detected as noisy by both the noise detection
stages are considered for noise removal by the fuzzy filtering process, which utilizes the direction based weighted median
to construct fuzzy membership function, which is the main contributing factor in noise removal and detail preservation. Extensive
experimentation shows that the proposed technique performs significantly better than state-of-the-art filters based on peak
signal-to-noise ratio, structural similarity index measure and subjective evaluation criteria. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|