Reconstruction performances of curvelet transform for magnetic resonance images

Reconstruction performances of transforms are deeply associated with image processing, scientific computing and computer vision. This research focuses on the performance of Curvelet Transform for Magnetic Resonance Images. The main outcome of this technique includes the removal of non-homogeneous noise using Curvelet based de-noising methods. Curvelet Transform belongs to the family of directional Wavelets. Curvelet Transform not only contains translations, dilations but also the rotations, which can enhance the reconstruction of curve objects. This research involves multi-scale reconstruction of objects with edge discontinuities. Experimental results show that Curvelet Transform has superior reconstruction capability for an image with curve objects. Another phase of this research covers the segmentation of de-noised images using Fuzzy C-Means Clustering (FCM) algorithm. The cluster formation in FCM algorithm is based on the Euclidian distance between pixels with similar intensities. Experimental results show that segmentation of reconstructed images is adversely affected by the noise bursts.