A Hybrid Prior-Based Framework for Infrared Image Enhancement Towards Reliable Scene Interpretation

Prior-Guided Histogram Equalization for Tunnel Image Enhancement Under Non-Uniform Illumination

Non-uniform illumination in tunnel environments severely degrades image quality, posing substantial challenges to visual monitoring and intelligent transportation systems. While histogram equalization (HE) remains prevalent due to its computational simplicity, its non-linear pixel transformations frequently induce over-enhancement, artifacts, and structural distortions. This paper proposes Prior-Guided Histogram Equalization (PGHE), a lightweight enhancement framework that integrates conventional HE with Retinex-based illumination priors. Within the Retinex decomposition paradigm, PGHE constructs a contrast illumination map from the ratio between the HE-enhanced image and the original input. A Prior Correction Module (PCM) subsequently refines this map via relative total variation regularization, thereby restoring spatial coherence and alleviating local discontinuities introduced by HE. The corrected map is then applied to the original image to obtain the final enhanced result. Extensive evaluation on the LOL low-light benchmarks and a proprietary tunnel dataset comprising 247 real-world frames shows that PGHE offers favorable trade-offs among contrast enhancement, structural fidelity, and brightness preservation: it is particularly strong in brightness preservation and Entropy, while its PSNR/SSIM on LOL and its NIQE on the tunnel dataset are comparable to, but not always the best among, the compared methods. Furthermore, the proposed PCM functions as a plug-in module that improves existing HE variants with measurable gains in Structural Similarity and perceived naturalness at a modest cost in Absolute Mean Brightness Error.

Depth-Prior Guided Learning for Infrared-Visible Image Registration and Fusion

Infrared and visible image fusion integrates complementary information to enhance scene perception. However, spatial misalignment caused by varying sensor poses and dynamic scenes often degrades the quality of fusion. Most existing approaches focus solely on fusing pre-registered images within the 2-D domain, neglecting the underlying geometric inconsistencies while lacking effective priors to ensure cross-modal structural consistency. To overcome these limitations, we propose a depth-prior guided registration and fusion framework. Firstly, we design a Depth-Guided Cross-Modal Attention (DCMA) module that incorporates depth geometric priors into linear attention, enabling robust and efficient cross-modal feature interaction. Secondly, our DCMA module serves as a unified component across both registration and fusion stages, enabling consistent depth-guided feature interaction for cross-modal alignment and complementary information aggregation. Finally, we introduce a lightweight Depth Quality Assessor (DQA) that generates a continuous quality score to adaptively interpolate between depth-guided and standard attention, maintaining robust performance when depth estimates are unreliable. Comprehensive evaluations on multiple benchmarks demonstrate that our method achieves state-of-the-art performance in both registration and fusion tasks, validating the effectiveness of depth-prior guided cross-modal learning for enhanced scene perception.

Universal Representation for Real-World Misaligned Infrared-Visible Image Fusion.

Infrared and visible image fusion is pivotal for robust visual perception across all weather conditions and scenes. Although deep learning-based methods have made notable progress, most either assume pre-aligned inputs or rely on implicit feature-space alignment, which fails to fundamentally address the amplification of registration errors and the loss of semantic structure in the fused results. To this end, we propose a universal representation and end-to-end framework for jointly registering and fusing unaligned infrared-visible image pairs, dubbed URMIF. Each image is mapped into modality-invariant (homogeneous) and modality-specific (heterogeneous) features: the invariant "structural skeleton" encodes geometry and semantics to stabilize alignment, while the specific "texture carrier" preserves thermal saliency and visible details to enable complementary fusion. Therefore, we propose a bi-directionally coupled registration-fusion module. This module performs hierarchical deformation estimation from coarse to fine, effectively mitigating visual mismatches caused by complex parallax in real-world scenes. Within this framework, the fusion component acts as the "evaluator" of registration, providing feedback regularization to update the deformation and suppress error accumulation. Furthermore, we introduce a dominant-plane prior as a scene-level constraint, seeding stable global and patch-wise homographies and reconciling cross-modal detail conflicts, to reinforce geometric consistency and semantic reliability. We also release a large-scale dataset comprising 1,500+ unaligned infrared/visible pairs with registration ground truth, spanning diverse illumination conditions and fields of view. Based on this dataset and additional benchmarks, extensive experiments validate that our framework achieves robust alignment and high-quality fusion on misaligned inputs, markedly reducing artifacts and improving the performance of downstream tasks such as detection and segmentation. Code and benchmark are available at https://github.com/ZengxiZhang/URMIF.

Infrared and Visible Image Fusion via Style-Based Recalibration and Edge Enhancement

Infrared and visible image fusion (IVIF) aims to preserve infrared thermal targets and visible structural textures in one informative image. Although recent attention-based methods improve cross-modal interaction, their post-fusion refinement remains limited in two aspects: modality-specific channel statistics are no longer explicitly exposed after feature mixing, and repeated attention-based aggregation can smooth spatial responses and weaken high-frequency visible details. To address these issues, this work proposes a lightweight end-to-end IVIF network with two complementary refinement modules. MSG carries out cross-modal style-based recalibration by making use of the joint mean and standard deviation of the two pre-fusion encoder features, so that first- and second-order pre-fusion modality statistics can guide post-fusion channel selection. DGM carries out edge enhancement by constructing a parameter-free Sobel detail prior from source images and learning only a lightweight residual modulation to perform restoration of high-frequency evidence. With only 80,160 trainable parameters, the proposed method achieves the best or tied-best value on three of seven standard fusion-quality metrics on FMB and four of seven on LLVIP, and ablation results further confirm the complementary effects of MSG and DGM.

DPSF-Net: A Dual-Prior Spatial-Frequency Network for Real-World Remote Sensing Image Dehazing

Real-world remote sensing image dehazing (RSID) remains challenging because atmospheric scattering, spatially non-uniform haze and colour distortion jointly degrade structural and spectral information. Most deep learning methods rely on RGB inputs and spatial-domain feature extraction, which limits their ability to separate global background haze from local surface details. Here, we propose DPSF-Net, a dual-prior spatial-frequency network built on MCAF-Net for real-world RSID. The network uses hazy RGB images and dark channel prior (DCP) maps as joint inputs, allowing physical degradation cues to guide end-to-end feature learning. A spatial-frequency residual interaction block introduces a FourierUnit branch into multi-directional spatial interaction to model large-scale haze components. A prior-guided feature attention module adaptively fuses prior and attention features to reduce colour shift and structural distortion. A selective kernel complementary fusion module screens multi-scale skip features through bidirectional residual complementary gating and selective kernel fusion. Extensive experiments demonstrate that DPSF-Net achieves state-of-the-art performance on the real-world RRSHID remote sensing image dehazing benchmark and remains competitive across multiple synthetic datasets. Moreover, the proposed method strikes a favourable balance among restoration quality, parameter count and computational complexity, supporting the effectiveness of dual-prior spatial-frequency modelling.

Prior-calibrated spatial-frequency Retinex decomposition for unsupervised low-light image enhancement