Utilizing the Structure of the Curvelet Transform with Compressed Sensing

TL;DR

This paper enhances compressed sensing image reconstruction by combining the curvelet transform with a modified approach that uses the Nyquist-Shannon theorem for low-frequency estimation, improving results on MRI and optical images.

Contribution

It introduces a novel method that modifies the sparsifying transformation to incorporate known low-frequency information, improving compressed sensing reconstruction.

Findings

Improved image quality in MRI and optical images.

Effective use of the Nyquist-Shannon theorem for low-frequency estimation.

Enhanced reconstruction accuracy with the modified transform.

Abstract

The discrete curvelet transform decomposes an image into a set of fundamental components that are distinguished by direction and size as well as a low-frequency representation. The curvelet representation is approximately sparse; thus, it is a useful sparsifying transformation to be used with compressed sensing. However, the low-frequency portion is seldom sparse. This manuscript presents a method to modify the redundant sparsifying transformation comprised of the wavelet and curvelet transforms to take advantage of this fact with compressed sensing image reconstruction. Instead of relying on sparsity for this low-frequency estimate, the Nyquist-Shannon theorem specifies a square region centered on the $0$ frequency to be collected, which is used to generate a blurry estimate. A Basis Pursuit Denoising problem is solved to determine the missing details after modifying the sparsifying transformation to take advantage of the known fully sampled region. Improvements in quality are shown on magnetic resonance and optical images.