Image Denoising Using Deep Learning: Convolutional Neural Network

This study will work on the denoising of images using Deep Learning techniques such as Convolutional Neural Networks (CNN). Digital images are playing a huge role in many aspects of our daily life like they are utilizing in satellite TV, traffic monitoring, signature approval etc., Because of impact of transmission channels and other factors, images are unavoidably being corrupted by noise during the process of image acquisition, compression and transmission resulting in the deformation and loss of image data.

 It has become an important task to remove noise from the image and restore high quality image for further image processing steps. This task will be performed by adding gaussian noise to the input image, and it is removed by using pre-trained denoised network ‘DnCNN’. Quality analysis is performed using the metrics like PSNR (Peak Signal to Noise Ratio), SSIM (Structural Similarity Index Measurement) and MSE (Mean Square Error).

Keywords: Image Denoising, Deep Learning, Convolutional Neural Network, Gaussian Noise.