Blind Deconvolution for Color Images Using Normalized Quaternion Kernels

2025-11-21

AI Summary

Conventional channel-wise decoupling in blind deconvolution of color images often incurs chromatic distortion due to loss of inter-channel correlations. Method: This paper proposes a quaternion-based blind deblurring framework that uniformly represents RGB images as quaternion signals. It introduces a quaternion fidelity term, employs a non-negative scalar kernel to model global blur, and utilizes three unconstrained quaternion kernels to capture complex cross-channel dependencies. A normalization mechanism ensures luminance consistency across channels. The optimization integrates quaternion convolution with differentiable kernel constraints. Results: Evaluated on real-world blurred datasets, the method significantly suppresses color artifacts and achieves superior PSNR and SSIM scores compared to state-of-the-art approaches, demonstrating both the effectiveness and necessity of explicitly modeling chromatic coupling in blind image deconvolution.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Non-convex Optimization
• Machine Learning: Kernel Methods

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

In this work, we address the challenging problem of blind deconvolution for color images. Existing methods often convert color images to grayscale or process each color channel separately, which overlooking the relationships between color channels. To handle this issue, we formulate a novel quaternion fidelity term designed specifically for color image blind deconvolution. This fidelity term leverages the properties of quaternion convolution kernel, which consists of four kernels: one that functions similarly to a non-negative convolution kernel to capture the overall blur, and three additional convolution kernels without constraints corresponding to red, green and blue channels respectively model their unknown interdependencies. In order to preserve image intensity, we propose to use the normalized quaternion kernel in the blind deconvolution process. Extensive experiments on real datasets of blurred color images show that the proposed method effectively removes artifacts and significantly improves deblurring effect, demonstrating its potential as a powerful tool for color image deconvolution.

Problem

• Blind deconvolution for color images using quaternion kernels
• Preserving color channel relationships during deblurring process
• Removing artifacts and improving deblurring for blurred color images

Innovation

• Uses quaternion fidelity term for color images
• Employs normalized quaternion kernel to preserve intensity
• Models color channel interdependencies with four kernels