Noise Reduction in Magnetic Resonance Images by Wavelet Transforms. An Application to the Study of Capillary Water Absorption in Sedimentary Rocks

Magnetic resonance imaging (MRI) is a powerful technique to study capillary water absorption kinetics in sedimentary rocks. However, the noise in the images can limit the correct identification and the quantitative measurement of the average height reached by the wetting front inside the porous material where imbibition occurs. Therefore, denoising methods can be applied to improve the image quality for a more accurate analysis, without the disadvantages of longer acquisition times. This study attempts to improve the signal-to-noise ratio of the images acquired by MRI on a sedimentary rock (Pietra di Lecce) using a waveletbased thresholding technique. The idea is to average some slightly different discrete wavelet tansforms (DWT), called e-decimated DWT, to define the stationary wavelet transforms (SWT).