期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2015
卷号：4
期号：12
页码：4310-4313
出版社：Shri Pannalal Research Institute of Technolgy
摘要：In image processing, restoration and noise reduction are expected to improve the quality of the image and the performance of quantitative image analysis. Adaptive median filtering techniques and the non-local means filtering algorithm are used to denoise image, corrupted by salt and pepper noise. Adaptive median filtering involves a two stage process, when the number of computations is very high and hence the simulation time increases with increase in the size of the corrupted image. DWT (Discrete Wavelet Transform) Denoising which exhibits good performance both in denoising and in restoration can be easily and effectively parallelized to exploit the full power of multi-core processor performance. The proposed Wavelet Denoising implementation based on the Fast Flow library achieves both close-to-ideal speedup and very good wall-clock execution figures. In Experimental result, PSNR provides better results when compared with the NLM.
关键词：Adaptive Median Filter; Salt and Pepper ; Noise; Median Filter; Principal Component Analysis Filter; ; Discrete Wavelet Transform(DWT); Peak Signal To Noise ; Ratio(PSNR).
