A Mesopic Lighting Evaluation Model Based on 1D Convolutional Neural Networks

Hung-Chung Li,Yennun Huang,Pei-Li Sun

doi:10.1109/icce-taiwan49838.2020.9258158

A Mesopic Lighting Evaluation Model Based on 1D Convolutional Neural Networks

Hung-Chung Li, Yennun Huang + Show 1 more

https://doi.org/10.1109/icce-taiwan49838.2020.9258158

Copy DOI

Publication Date: Sep 28, 2020

Citations: 2

Affiliation: Research Center for Information Technology Innovation, Academia Sinica, National Taiwan University of Science and Technology

#Color Gamut Volume #1D Convolutional Neural Networks + Show 8 more

Abstract
Full-Text PDF
Similar Papers

Abstract

In the study, a mesopic lighting evaluation model that can be used to predict S/P ratio, mesopic luminance and color gamut volume of a white LED spectrum is proposed. The result shows that the one-dimensional convolutional neural network describes all of the indices well especially for mesopic luminance and color gamut volumes with a high coefficient of determination. In contrast, the performance of the S/P ratio is slightly unfavorable.

Full Text