الفهرس Only 14 pages are availabe for public view
 |  | from | 114 |  |  |
| | | from | 114 | | |
Abstract
Image noise reduction is one of the most critical problems in the current signal processing researches. the application of the image noise reductions versatile including , restoration of the original image in remote signal transfer and refinement the low resolution magnetic resonance images(MRI)
In this thesis, two methods are utilized to solve the denoising problem. first method is depending on the Neural Networks and its effects of keeping the features of the original signals. Second is depending on Wavelets and its effects of noise suppression and image restoration
In this study , both techniques are outlined; their mathematical concepts, physical meaning, and the related algorithms needed for processing are discussed. Numerical examples for both methods to check the effect of noise reduction are illustrated
During the search, a new recurrent method for using the Neural Networks in image noise reduction is presented.
Also a novel algorithm “Alternating Directions Wavelet Denoise (ADWD)” is proposed to improve the discrete Wavelet Transform (DWT) effect in denosing process. Numerical results show that the suggested methods are effective in noise reduction. Also, a comparison is set between a classic denoising algorithm”Wiener Filter” classic wavelet denoise ,and ADWD.