Study of the reconstruction of reflectance curves based on tristimulus values: Comparison of methods of optimization

This article presents and compares different methods applied to the reconstruction of reflectance curves. These are optimization methods such as the simplex method, simulated annealing, genetic algorithms, neural networks, and Hawkyard's method, which is specific to colorimetry. Improvements have been made to these methods to maximize their application, notably to ensure a convergence toward the global optimum by avoiding being trapped in local optima and to obtain physically viable reflectance curves. Objective and subjective criteria were used to compare these methods; the study used textile samples (cotton fibers), with a panel of 15 samples covering the CIELAB L*a*b* space.