Mizuno, AkitoTanaka, KojiroMikawa, MasahikoFujisawa, MakotoJean-Marie NormandMaki SugimotoVeronica Sundstedt2023-12-042023-12-042023978-3-03868-218-91727-530Xhttps://doi.org/10.2312/egve.20231307https://diglib.eg.org:443/handle/10.2312/egve20231307Recently, a wide range of media devices have been developed, with increasing emphasis on activities related to media art using natural elements. Although numerous studies have been conducted to harmonize media devices with natural landscapes, there is research on a display device using grass, termed artificial grass display, and its foundational element, artificial grass pixel. The artificial grass pixel employs the mechanism of pin displays and two types of grass to dynamically control the color of the grass. Moreover, studies are being conducted on the method of color scale mapping for the artificial grass pixel. However, there is a problem in experiments in the real space conditions due to the experimental overhead of acquiring a lot of data for color scale mapping. In order to solve this problem, this study aims to replicate and evaluate the artificial grass pixel in a virtual space using Computer Graphics (CG) technology. In this paper, we propose a novel method to estimate characteristic parameters of CG models based on a genetic algorithm to minimize the color difference from a real model. As a result, by performing calibration using multiple sets of real data, it was confirmed that the simulated grass pixel demonstrated color and color scale evaluation results aligned based on the color difference with the grass pixel in the real space.Attribution 4.0 International LicenseCCS Concepts: Computing methodologies → Modeling and simulationComputing methodologies → Modeling and simulation10.2312/egve.202313071-99 pages