Learning Priors in High-frequency Domain for Inverse Imaging Reconstruction

TL;DR

This paper introduces a novel high-frequency transform-guided denoising autoencoder as a prior for inverse imaging problems, improving reconstruction quality in highly under-sampled MRI and CT scans.

Contribution

It proposes a new multi-profile high-frequency prior model integrated into iterative reconstruction, enhancing detail recovery in inverse imaging tasks.

Findings

Effective in reconstructing feature details in under-sampled MRI and CT.

Outperforms state-of-the-art methods in preliminary tests.

Utilizes high-frequency components for improved semantic information capture.

Abstract

Ill-posed inverse problems in imaging remain an active research topic in several decades, with new approaches constantly emerging. Recognizing that the popular dictionary learning and convolutional sparse coding are both essentially modeling the high-frequency component of an image, which convey most of the semantic information such as texture details, in this work we propose a novel multi-profile high-frequency transform-guided denoising autoencoder as prior (HF-DAEP). To achieve this goal, we first extract a set of multi-profile high-frequency components via a specific transformation and add the artificial Gaussian noise to these high-frequency components as training samples. Then, as the high-frequency prior information is learned, we incorporate it into classical iterative reconstruction process by proximal gradient descent technique. Preliminary results on highly under-sampled magnetic resonance imaging and sparse-view computed tomography reconstruction demonstrate that the proposed method can efficiently reconstruct feature details and present advantages over state-of-the-arts.