Abstract:Remote sensing images are essential for many applications of the Earth's sciences, but their quality can usually be degraded due to limitations in sensor technology and complex imaging environments. To address this, various remote sensing image deblurring methods have been developed to restore sharp and high-quality images from degraded observational data. However, most traditional model-based deblurring methods usually require predefined hand-crafted prior assumptions, which are difficult to handle in complex applications. On the other hand, deep learning-based deblurring methods are often considered as black boxes, lacking transparency and interpretability. In this work, we propose a new blind deblurring learning framework that utilizes alternating iterations of shrinkage thresholds. This framework involves updating blurring kernels and images, with a theoretical foundation in network design. Additionally, we propose a learnable blur kernel proximal mapping module (KPMM) to improve the accuracy of the blur kernel reconstruction. Furthermore, we propose a deep proximal mapping module in the image domain, which combines a generalized shrinkage threshold with a multiscale prior feature extraction block. This module also incorporates an attention mechanism to learn adaptively the importance of prior information, improving the flexibility and robustness of prior terms, and avoiding limitations similar to hand-crafted image prior terms. Consequently, we design a novel multiscale generalized shrinkage threshold network (MGSTNet) that focuses specifically on learning deep geometric prior features to enhance image restoration. Experimental results on real and synthetic remote sensing image datasets demonstrate the superiority of our MGSTNet framework compared to existing deblurring methods.

Image Smoothing via Multiscale Global Perception

Taming Vector-Wise Quantization for Wide-Range Image Blending with Smooth Transition

Deep Scale-Aware Image Smoothing

Image smoothing via unsupervised learning

Image Smoothing Based on Image Decomposition and Sparse High Frequency Gradient

Image smoothing based on global sparsity decomposition and a variable parameter

Learning Explicit Smoothing Kernels for Joint Image Filtering

Image smoothing method based on global gradient sparsity and local relative gradient constraint optimization

Contrastive Semantic-Guided Image Smoothing Network

Structure-preserving image smoothing with semantic cues

Pixel-Level Non-local Image Smoothing With Objective Evaluation

Real-time Image Smoothing via Iterative Least Squares

A Parameter Adaptive Method for Image Smoothing

Multi-scale network for single image deblurring based on ensemble learning module

GSmooth: Certified Robustness against Semantic Transformations via Generalized Randomized Smoothing

Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring

Multi-scale Grid Network for Image Deblurring with High-frequency Guidance

Multiscale Structure Guided Diffusion for Image Deblurring

An Enhance Relative Total Variation with BF Model for Edge-preserving Image Smoothing

A Multi-scale Generalized Shrinkage Threshold Network for Image Blind Deblurring in Remote Sensing

A Multiscale Generalized Shrinkage Threshold Network for Image Blind Deblurring in Remote Sensing