FISTA-Net: Learning A Fast Iterative Shrinkage Thresholding Network for Inverse Problems in Imaging

Abstract

Inverse problems are essential to imaging applications. In this paper, we propose a model-based deep learning network, named FISTA-Net, by combining the merits of interpretability and generality of the model-based Fast Iterative Shrinkage/Thresholding Algorithm (FISTA) and strong regularization and tuning-free advantages of the data-driven neural network. By unfolding the FISTA into a deep network, the architecture of FISTA-Net consists of multiple gradient descent, proximal mapping, and momentum modules in cascade. Different from FISTA, the gradient matrix in FISTA-Net can be updated during iteration and a proximal operator network is developed for nonlinear thresholding which can be learned through end-to-end training. Key parameters of FISTA-Net including the gradient step size, thresholding value and momentum scalar are tuning-free and learned from training data rather than hand-crafted. We further impose positive and monotonous constraints on these parameters to ensure they converge properly. The experimental results, evaluated both visually and quantitatively, show that the FISTA-Net can optimize parameters for different imaging tasks, i.e. Electromagnetic Tomography (EMT) and X-ray Computational Tomography (X-ray CT). It outperforms the state-of-the-art model-based and deep learning methods and exhibits good generalization ability over other competitive learning-based approaches under different noise levels.

arXiv comment11 pages;

References in corpus (14)

• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)
• Low Dose CT Image Denoising Using a Generative Adversarial Network with Wasserstein Distance and Perceptual Loss
• Fast Image Recovery Using Variable Splitting and Constrained Optimization
• MoDL: Model Based Deep Learning Architecture for Inverse Problems
• On instabilities of deep learning in image reconstruction - Does AI come at a cost?
• Learned Primal-dual Reconstruction
• A Review of Convolutional Neural Networks for Inverse Problems in Imaging
• Solving ill-posed inverse problems using iterative deep neural networks
• Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?
• Deep Residual Learning for Compressed Sensing CT Reconstruction via Persistent Homology Analysis
• Learning with Known Operators reduces Maximum Training Error Bounds
• A Recursive Born Approach to Nonlinear Inverse Scattering
• Multi-frequency Electromagnetic Tomography for Acute Stroke Detection Using Frequency Constrained Sparse Bayesian Learning

Cited by in corpus (10)

• Physics-/Model-Based and Data-Driven Methods for Low-Dose Computed Tomography: A survey
• An Interpretable MRI Reconstruction Network with Two-grid-cycle Correction and Geometric Prior Distillation
• Nest-DGIL: Nesterov-optimized Deep Geometric Incremental Learning for CS Image Reconstruction
• A Compound Gaussian Least Squares Algorithm and Unrolled Network for Linear Inverse Problems
• Convolutional Sparse Coding Fast Approximation with Application to Seismic Reflectivity Estimation
• Optimal MRI Undersampling Patterns for Ultimate Benefit of Medical Vision Tasks
• SINET: Sparsity-driven Interpretable Neural Network for Underwater Image Enhancement
• JotlasNet: Joint Tensor Low-Rank and Attention-based Sparse Unrolling Network for Accelerating Dynamic MRI
• WARP-LCA: Efficient Convolutional Sparse Coding with Locally Competitive Algorithm
• An Unsupervised Learning Method for Radio Interferometry Deconvolution