Compressive Sensing of Image Reconstruction Based on Shearlet Transform

摘要

Compressive sensing (CS) is a radical new way of sampling signal at a sub-Nyquist rate, acquires a signal of interest indirectly by correcting a very limited number of its "projections", and the signal can be exactly reconstructed from these "projections", this new signal acquisition paradigm has revolutionized the way digital data are traditionally acquired. Generally, researchers always use orthogonal wavelet as sparse basis, but it fails to provide an optimal sparse representation for images that contain texture details. In this paper, we use shearlet which is a new directional multiresolution transform, can efficiently represent the directional information of images, meanwhile, using the RecPF of [4] as the reconstruction algorithm. The experimental results indicate that the quality of reconstructed image is improved and obtain better performance of PSNR.