Brain MR image denoising based on wavelet transform

Images must be clear and noise free, in order to achieve better accuracy in classification results of brain tumor from magnetic resonance imaging (MRI). But in the process of collection of medical images, the picture gets noisy, inadvertently. Deletion of noise from images is known as wavelet shrinkage or thresholding. In this work, an ingenious and compatible method is proposed for the valuation of thresholding parameters, hinge on the information of wavelet coefficients. For the better illustration of the process brain MRI was introduced with Gaussian noise at the different level of variances and then denoised using Wavelet Transform with coding in MATLAB. The same procedure was repeated to denoise three brain MR Images with the brain tumor. Proposed method helps in embellished off the edge and texture details of the images. The image quality of brain MR images is assessed in terms of peak signal-to-noise ratio (PSNR). Experimental results represent that this method attain preferable denoised image with improved PSNR.