Title :
Adaptive Salt-&-Pepper Noise Removal: A Function Level Evolution based Approach
Author :
Li, Jie ; Huang, Shitan
Author_Institution :
Xi´´an Microelectron. Technol. Inst., Xi´´an
Abstract :
Noise removal is an important procedure before ongoing further image processing. In this paper, a two-stage adaptive noise removal scheme is proposed. The first stage of the scheme is noise detection; the second stage is functional level evolution based noise cancellation. This cancellation only operates on the noise candidates detected in the first stage. Additionally, the proposed method acquires local information to suppress noise from the noisy pixelspsila neighbors rather than from original images. These features enable the scheme to preserve more image details with less computation effort, and make it able to be the basic frame for EHW based implementation of on-line adaptive impulse noise removal system. Simulation results show that the proposed approach provides at least equal performance to the adaptive median filter, and can cope with the corrupted images with up to 70% noise density.
Keywords :
adaptive filters; evolutionary computation; image denoising; impulse noise; median filters; EHW; adaptive median filter; adaptive salt-&-pepper noise removal; function level evolution; image processing; noise cancellation; noise detection; online adaptive impulse noise removal system; two-stage adaptive noise removal scheme; Adaptive filters; Adaptive systems; Circuit noise; Filtering; Hardware; NASA; Noise cancellation; Noise level; Pixel; Samarium; Adaptive Noise Removal; Evolvable Hardware; Function level evolution; Salt-&-Pepper Noise;
Conference_Titel :
Adaptive Hardware and Systems, 2008. AHS '08. NASA/ESA Conference on
Conference_Location :
Noordwijk
Print_ISBN :
978-0-7695-3166-3
DOI :
10.1109/AHS.2008.27