High fidelity colour reproduction under non linear illumination conditions

Abstract

Accurate colour reproduction remains a critical challenge in imaging systems operating under non-linear illumination, where interactions among light sources, material reflectance, and sensor response deviate from ideal radiometric assumptions. Real-world illumination is rarely uniform, varying spatially and spectrally and causing colour shifts, distortions, and inconsistencies in captured images. Conventional colour correction methods often underperform in complex environments because of their reliance on linear models and limited adaptability. This paper presents a novel framework for high-fidelity colour reproduction under non-linear illumination, integrating adaptive radiometric transformation with context-aware colour constancy modelling. The proposed approach employs a dynamic correction mechanism to estimate illumination-induced distortions and compensate for them through iterative refinement. By combining nonlinear response modelling, localized colour mapping, and adaptive normalization, the framework restores chromatic consistency while preserving perceptual realism. Experimental results under diverse illumination conditions demonstrate improved colour accuracy, visual consistency, and robustness, making the framework suitable for advanced imaging applications..