An LED high-fidelity color gamut conversion model based on convex optimization and DE2000 chromatic aberration
基于凸优化与DE2000色差的LED高保真色域转换模型

摘要

针对 LED 显示中广色域视频源到有限色域 RGB 屏的转换失真及多基色系统转换中的欠定与非负约束矛盾，提出基于 CIE DE2000 色差优化的色域转换模型。广色域映射采用 “线性转换 - 凸约束 - 局部优化” 模型，将 BT.2020 转换至 CIE-XYZ 后，对超色域点沿亮度轴投影并在凸约束下优化色差；多基色转换设计 “非负最小二乘 + 凸色域投影” 算法，确保物理可行并优化域外解。结果显示，BT2020→RGB 转换平均 DE2000 色差降至 3.24（较传统方法降低 44.3%），RGBV→RGBCX 转换 95.6% 的点位于目标色域内（平均色差 3.05）。该模型为 LED 显示高保真色域转换提供了有效框架。

Abstract

To address the distortion in converting wide color gamut video sources to limited-color-range RGB displays and the trade-off between underdetermination and non-negativity constraints in multi-primary system conversion, this study proposes a color gamut conversion model optimized for chromatic aberration based on CIE DE2000. The wide color gamut mapping employs a "linear transformation-convex constraint-local optimization" framework: after converting BT.2020 to CIE-XYZ, super-color-gamut points are projected along the brightness axis and chromatic aberration is optimized under convex constraints. For multi-primary conversion, we design an algorithm combining "non-negative least squares with convex color gamut projection" to ensure physical feasibility while optimizing out-of-domain solutions.
Results demonstrate that the BT2020-to-RGB conversion achieves an average DE2000 chromatic aberration reduction of 3.24 (44.3% lower than traditional methods), while the RGBV-to-RGBCX conversion places 95.6% of points within the target color gamut (average chromatic aberration 3.05). This model provides an effective framework for high-fidelity color gamut conversion in LED displays.