Image Denoising for AWGN Corrupted Image Using OWT and Thresholding

In this work, review of various well-known algorithms for image denoising is carried out and their performances with their methodologies are comparatively assessed. A new algorithm based on the orthonormal wavelet transform OWT is developed. In this work images corrupted by AWGN are denoised. Simulation results shows that proposed method using Orthonormal wavelets for different values of noise Standard Deviation s in dB outperforms other available methods. Also Coiflet Wavelet performs better than Symlet, Haar and Daubechies wavelets. The proposed Orthonormal wavelet transform OWT method has minimum Mean Square and highest PSNR with Coif let wavelets. Simulation results shows that denoised image is 98.29 similar for 5 dB noise standard deviation and 84.42 similar for 30 dB noise standard deviation. The proposed method has 1.35 dB and 4 improvement for s =10 dB, 2.08 dB and 7 improvement for s =20 dB and 2.26 dB and 9 improvement for s =30 dB as compared to denoising with two thresholds for edge detection 1 . Shruti Badgainya | Prof. Pankaj Sahu | Prof. Vipul Awasthi "Image Denoising for AWGN Corrupted Image Using OWT & Thresholding" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-2 | Issue-6 , October 2018, URL: http://www.ijtsrd.com/papers/ijtsrd18338.pdf