Color Space Mapping and Correction Study Based on Nonlinear Mapping Function and Differential Evolutionary Algorithm

Abstract

This paper proposes a color mapping and calibration model based on nonlinear mapping functions and optimization algorithms, focusing on the implementation methods for different gamut conversions and LED color calibration. First, a nonlinear mapping function with 21 parameters is constructed. Combined with the total loss function consisting of color difference, gamut constraints, and smoothness loss, the differential evolutionary algorithm is employed to map the BT.2020 gamut to the gamut of ordinary displays, effectively reducing color loss. Secondly, a nonlinear mapping function with 45 parameters is designed, and the differential evolutionary algorithm is also used to optimize the total loss function, realizing the mapping from four-channel to five-channel color spaces while taking both color accuracy and constraint conditions into account. Finally, the coefficient matrix calibration principle and quadratic polynomial mapping scheme are adopted, and the color calibration of LED displays is completed by minimizing the total loss function. This model can achieve accurate conversion of different gamut and color calibration of display devices, with the advantages of efficient parameter optimization, small color loss, and smooth mapping.