Removal of salt-and-pepper noise for X-ray bio-images using pixel-variation gain factors. (October 2018)

• Record Type: Journal Article
• Authors: Lu, Ching-Ta; Chen, Mu-Yen; Shen, Jun-Hong; Wang, Ling-Ling; Hsu, Chih-Chan
• Is Part Of: Computers & electrical engineering. Volume 71(2018)
• Page Start: 862
• Page End: 876
• Publication Date: 2018-10
• Subjects: Noise removal -- Salt-and-pepper noise -- Bio-image denoising -- Gain factor -- Majority and variation adaptation; SAP salt-and pepper -- GLV gray level variation -- MDBUTM modified decision based un-symmetric trimmed median -- DWM directional-weighted-median -- PSNR peak signal-to-noise ratio -- MSSIM mean structural similarity; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2017.08.012
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Salt-and-pepper noise can be efficiently removed by weighting the neighbor pixels. Employ the weighted value of non-extreme pixels restore the corrupted pixel well. Majority decision is beneficial for X-ray bio-image restoration. Adaptation of the number of noise-free pixel and pixel variation is effective. Abstract: Salt-and-pepper (SAP) noise may degrade the bio-images at the time of capturing or transmission. In this paper, we propose a gain factor adapted by a noise-free pixel number and pixel variation for the removal of the impulse noise. All pixels with a non-extreme gray level value are sorted in an ascending order and are grouped according to the gray level variation of pixels in an analysis window. The distribution ratio and median value of each group are computed to determine the values of the gain factors. They are multiplied with the median value of each group to obtain a weighted value which is employed to replace the center pixel with an extreme gray level value, enabling noise-corrupted pixels to be reconstructed. Experimental results show that the proposed approach can effectively remove SAP noise from a corrupted bio-image for various noise corruption densities; meanwhile, the denoised bio-image is not blurred.