QRetinex-Net: Quaternion-Valued Retinex Decomposition for Low-Level Computer Vision Applications

TL;DR

This paper introduces Quaternion Retinex-Net, a novel approach using quaternion algebra for better low-light image decomposition, improving color fidelity, noise reduction, and reflectance stability in various computer vision tasks.

Contribution

The paper presents the first quaternion-valued Retinex model, addressing key flaws of classic methods and demonstrating improved performance in low-light image applications.

Findings

Achieved 2-11% performance gains over leading methods.

Enhanced color fidelity and reduced noise in low-light images.

Introduced the Reflectance Consistency Index for stability assessment.

Abstract

Images taken in low light often show color shift, low contrast, noise, and other artifacts that hurt computer-vision accuracy. Retinex theory addresses this by viewing an image S as the pixel-wise product of reflectance R and illumination I, mirroring the way people perceive stable object colors under changing light. The decomposition is ill-posed, and classic Retinex models have four key flaws: (i) they treat the red, green, and blue channels independently; (ii) they lack a neuroscientific model of color vision; (iii) they cannot perfectly rebuild the input image; and (iv) they do not explain human color constancy. We introduce the first Quaternion Retinex formulation, in which the scene is written as the Hamilton product of quaternion-valued reflectance and illumination. To gauge how well reflectance stays invariant, we propose the Reflectance Consistency Index. Tests on low-light crack inspection, face detection under varied lighting, and infrared-visible fusion show gains of 2-11 percent over leading methods, with better color fidelity, lower noise, and higher reflectance stability.