Quaternion Multi-focus Color Image Fusion

2025-05-05

AI Summary

To address color distortion and texture detail loss in multi-focus color image fusion, this paper proposes, for the first time, an end-to-end quaternion-domain all-in-focus fusion framework. Methodologically, we formulate a quaternion sparse decomposition model, design a base-detail hierarchical fusion strategy, and incorporate a structural similarity–driven adaptive local inpainting mechanism to jointly enhance focus detection accuracy, structural fidelity, and spatial consistency. Unlike conventional RGB/HSV-domain approaches, our quaternion-based formulation avoids chromatic distortion induced by channel-wise decoupling, enabling joint modeling of color and texture. Extensive experiments on multiple benchmark datasets demonstrate consistent improvements: average gains of 2.1 dB in PSNR, 0.032 in SSIM, and 0.028 in FSIM—particularly pronounced in highly saturated and complex-textured regions.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Multi-focus color image fusion refers to integrating multiple partially focused color images to create a single all-in-focus color image. However, existing methods struggle with complex real-world scenarios due to limitations in handling color information and intricate textures. To address these challenges, this paper proposes a quaternion multi-focus color image fusion framework to perform high-quality color image fusion completely in the quaternion domain. This framework introduces 1) a quaternion sparse decomposition model to jointly learn fine-scale image details and structure information of color images in an iterative fashion for high-precision focus detection, 2) a quaternion base-detail fusion strategy to individually fuse base-scale and detail-scale results across multiple color images for preserving structure and detail information, and 3) a quaternion structural similarity refinement strategy to adaptively select optimal patches from initial fusion results and obtain the final fused result for preserving fine details and ensuring spatially consistent outputs. Extensive experiments demonstrate that the proposed framework outperforms state-of-the-art methods.

Problem

• Integrating partially focused color images into one all-in-focus image
• Handling color information and intricate textures in complex scenarios
• Improving fusion quality using quaternion domain techniques

Innovation

• Quaternion sparse decomposition for focus detection
• Quaternion base-detail fusion for structure preservation
• Quaternion similarity refinement for detail consistency