Researchers Arathi Sankar and Jayakumar E P have proposed a novel visible and near-infrared (NIR) image fusion framework designed to address the significant information loss that occurs in visible images under low-light environments. Published in the journal Multimedia Tools and Applications, the study introduces a method that integrates a modified bright prior (MBP) enhancement model with edge-preserving filtering techniques. The core objective of this framework is to produce single, information-rich images that preserve structural details, textures, and edge information while simultaneously improving overall visibility in challenging lighting conditions.
The research addresses a specific technical challenge in image processing: while visible images often suffer from degraded quality in low-light settings, near-infrared images tend to preserve structural details more effectively. To leverage these complementary strengths, the proposed framework employs weighted guided image filtering (WGIF) and fast bilateral filtering (FBF). These filtering techniques are used to decompose the input images into distinct base and detail parts, which facilitates the adaptive fusion of information from both visible and NIR modalities.
The key contribution of the study is the combination of MBP-based illumination enhancement and a contrast-aware fusion strategy. This component is specifically designed to retain fine textures and salient features that might otherwise be lost during the fusion process. By combining MBP-based illumination enhancement with a contrast-aware fusion strategy, the method aims to efficaciously preserve the structural integrity of the image while boosting its contrast.
The performance of the proposed method was assessed through both visual and quantitative evaluations against existing image fusion techniques. For the objective comparison, the researchers utilized six no-reference image quality metrics. The evaluation focused on how well the method could enhance contrast and preserve details under low-light conditions.
The results demonstrated that the proposed approach achieves superior fusion quality compared to state-of-the-art methods. The generated images exhibited enhanced contrast, improved detail preservation, and superior visual quality, particularly in scenarios where lighting conditions were challenging. Both qualitative and quantitative analyses validated the effectiveness of the framework, showing that it outperforms existing fusion frameworks, including recent deep learning-based methods, in terms of metric scores and visual appearance.
The work was published in Volume 85, article number 840, of the journal Multimedia Tools and Applications in 2026.
Subject of Research: Visible and near-infrared image fusion for low-light enhancement
Article Title: Low-light image enhancement in visible and near-infrared image fusion using modified bright prior
Article References: Sankar P, A., & E P, J. (2026). Low-light image enhancement in visible and near-infrared image fusion using modified bright prior. Multimedia Tools and Applications, 85(11), Article 840. https://doi.org/10.1007/s11042-026-21997-x
Image Credits: AI Generated
DOI: 10.1007/s11042-026-21997-x
Keywords: image fusion, low-light enhancement, near-infrared imaging, modified bright prior, image processing, multimedia tools, edge-preserving filtering, contrast enhancement, Low-light, image, enhancement, visible
News Source: Denise Maddox. (November 7, 2026). New Fusion Framework Enhances Low-Light Visible and NIR Images. Scienmag.



