CEMMO: A Multi-Channel LED Mapping Optimization Algorithm for Precise Color Reproduction

Abstract

Traditional LED display systems employing RGB three-channel structures face significant limitations in color gamut coverage due to their triangular primary color configurations, resulting in color deviation, transition distortion, and color compression issues that cannot meet the stringent visual fidelity requirements of emerging applications. This research proposes extending the input from traditional RGB to a four-channel RGBV structure incorporating luminance information, while introducing a five-channel emission mechanism RGBCX at the output end. By inserting auxiliary channels C (cyan) and X (yellow) between green-blue and red-green regions, this approach effectively covers mixed color regions outside the RGB triangle. We construct a 4 × 5 mapping matrix W and design a differentiable optimization objective function that jointly considers color fidelity (CIEDE2000 color difference), color gamut coverage (based on five-primary convex hull area), channel balance (variance minimization), energy conservation, and non-negativity constraints. A staged optimization algorithm framework combining non-negative least squares (NNLS) and regularized gradient descent is proposed to efficiently solve the multi-objective mapping weight matrix. This approach addresses the critical demands of ultra-high-definition video (UHD), augmented reality (AR), virtual reality (VR), and human-computer interaction applications for enhanced color vividness and spatial consistency. Experimental validation demonstrates that CEMMO achieves 23.6% improvement in color reproduction accuracy, 19.4% increase in color gamut volume, while reducing energy consumption by 5.8%. The proposed methodology establishes an efficient transformation framework from four-channel input signals to five-channel display outputs.