Breaking Degradation Coupling: A Structural Entropy Guided Decoupled Framework and Benchmark for Infrared Enhancement

2026-04-24

AI Summary

This work addresses the challenges of gradient interference and parameter competition arising from coupled composite degradations in thermal infrared image enhancement. To this end, we propose SEGD, a structure entropy-guided decoupling framework that decomposes composite degradation into independent sub-processes and models each via degradation-specific residual modules. For the first time, a structure entropy criterion is introduced to guide adaptive fusion of multi-path features, enabling interpretable and fine-grained enhancement. Additionally, we construct the first real-world benchmark for low-illumination nighttime thermal infrared image enhancement. Extensive experiments demonstrate that SEGD outperforms state-of-the-art methods on this new benchmark, achieving superior efficiency and structural fidelity with fewer parameters.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Security and Privacy: Privacy-enhancing technologies

Abstract

Thermal infrared image enhancement aims to restore high-quality images from complex compound degradations. Existing all-in-one approaches typically employ a single shared backbone to handle diverse degradations, which causes gradient interference and parameter competition. To address this, we propose a Structural Entropy-Guided Decoupled (SEGD) Framework. Unlike unified modeling paradigms, SEGD decomposes compound degradations into independent sub-processes and models them in a divide-and-conquer manner through Degradation-Specific Residual Modules (DRMs). Each DRM focuses on residual estimation for a specific degradation, enabling task decoupling while remaining jointly trainable, which mitigates parameter contention. A Degradation-Aware Evidential Network further estimates degradation type and intensity, providing priors that adaptively regulate DRM restoration strength. To handle compound cases, DRMs are composed in varying orders to form multiple restoration paths, from which the most informative features are aggregated under a structural-entropy criterion, yielding decoder-ready representations with structural fidelity and degradation awareness. Integrating divide-and-conquer restoration, evidential perception, and entropy-guided adaptation, SEGD achieves fine-grained and interpretable enhancement. We also construct a nighttime TIR benchmark for evaluation under real low-light conditions. Experimental results demonstrate that SEGD surpasses state-of-the-art methods while achieving higher efficiency with fewer parameters.

Problem

• infrared enhancement
• compound degradations
• gradient interference
• parameter competition
• thermal infrared image

Innovation

• Decoupled Framework
• Structural Entropy
• Degradation-Specific Residual Modules
• Evidential Network
• Infrared Image Enhancement