Chromatic-aware gradient-domain guided filtering for halo-free multi-exposure fusion

Multi-exposure image fusion (MEF) combines differently exposed images of the same high-dynamic-range scene to produce a high-quality low-dynamic-range image. However, existing MEF methods often struggle to preserve fine details in extremely bright and dark regions and may introduce halo artifacts and edge distortions around high-contrast boundaries. To address these challenges, we propose a chromatic-aware MEF framework with dynamic parameter selection and two-scale decomposition. Specifically, we incorporate RGB chromatic gradient information into edge-aware weight refinement to enhance cross-channel structural consistency and suppress halo artifacts. A global structure-adaptive parameter selection strategy based on percentile-based gradient aggregation dynamically determines the filtering radius and regularization parameter, thereby eliminating the need for manual parameter tuning. In addition, a two-scale coarse–residual decomposition separately preserves global luminance balance and fine-scale textural details. Extensive comparisons with eight state-of-the-art methods demonstrate superior perceptual quality, while quantitative evaluations confirm the effectiveness of the proposed framework. The framework maintains low computational overhead, with an average processing time of 0.77 s per image on a standard CPU, making it suitable for resource-constrained imaging applications.

Authors

Institutions

Publication Details

Journal
Journal of Electronic Imaging
Published
2026-11-14
DOI
https://doi.org/10.1117/1.jei.35.6.063009
Primary Topic
Advanced Image Fusion Techniques
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Chromatic-aware gradient-domain guided filtering for halo-free multi-exposure fusion

Rahul Dinesh Kumar
Journal of Electronic Imaging
Advanced Image Fusion Techniques
article

Chromatic-aware gradient-domain guided filtering for halo-free multi-exposure fusion

Rahul Dinesh Kumar
article en

Abstract

Multi-exposure image fusion (MEF) combines differently exposed images of the same high-dynamic-range scene to produce a high-quality low-dynamic-range image. However, existing MEF methods often struggle to preserve fine details in extremely bright and dark regions and may introduce halo artifacts and edge distortions around high-contrast boundaries. To address these challenges, we propose a chromatic-aware MEF framework with dynamic parameter selection and two-scale decomposition. Specifically, we incorporate RGB chromatic gradient information into edge-aware weight refinement to enhance cross-channel structural consistency and suppress halo artifacts. A global structure-adaptive parameter selection strategy based on percentile-based gradient aggregation dynamically determines the filtering radius and regularization parameter, thereby eliminating the need for manual parameter tuning. In addition, a two-scale coarse–residual decomposition separately preserves global luminance balance and fine-scale textural details. Extensive comparisons with eight state-of-the-art methods demonstrate superior perceptual quality, while quantitative evaluations confirm the effectiveness of the proposed framework. The framework maintains low computational overhead, with an average processing time of 0.77 s per image on a standard CPU, making it suitable for resource-constrained imaging applications.

Journal of Electronic ImagingVol. 35(06)
University of Kerala (IN)
Advanced Image Fusion Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon