Image denoising using multiple wavelet decomposition with bicubic interpolation

With the advent of better computers and high computing speeds use of images and videos have drastically increased. Today images are a very integral part of our lives from the entertainment industry to medical sciences. In the field of medicine, image processing plays a very important role when it comes to medical imaging. Image processing is utilized to get a clear denoised image for clear and easy diagnostic study. With increase in the usage, need to compress the images and store as many as possible in limited spaces have thus become a necessity. The emphasis being on the ability to convert them back into clear crisp image with minimum noise when the need be. Compressing an image is significantly different than compressing raw binary data. General purpose compression programs can be used to compress images, but the result is less than optimal. This is because images have certain statistical properties which can be exploited by encoders specifically designed for them. Also, some of the finer details in the image can be sacrificed for the sake of saving a little more bandwidth or storage space. This also means that lossy compression techniques can be used in this area. In this work, noisy image is decomposed using three different DWT transforms (haar/db6/coif5). Input images are processed with various noise levels (20, 50 and 80 %) with both salt and pepper and Additive White Gaussian Noise (AWGN). Various error metrics such as Mean Squared Error (MSE), Peak Signal to Noise Ratio (PSNR), Root Mean Squared Error (RMSD), Mean Absolute Error (MAE), and Structural Similarity index Measure (SSIM) are computed and compared with other state of art methods for stability performance.