A quantitative comparative study of image denoising algorithms: Conventional vs Deep Learning algorithms

MRI examination is one of the medical supports to assess the structure and anatomy of vertebrae thoracal. The modality that can be used is low field MRI. The disadvantages are produces low signals and noise. If the signal is low and the noise is high then the SNR value is low. As long as the solution is to image denoising, we will discuss in this research paper an analytical comparative study of conventionnal image denoising approches using bilateral filter and BM3D On the one hand, methods that rely on deep learning, especially DnCNN.