19 December 2013 Curve fitting for standard lamp of spectral irradiance based on RBFNN
Author Affiliations +
Abstract
To reduce the uncertainty of dissemination, the models for standard lamp of spectral irradiance data are presented. We propose a divide-and-conquer RBF neural network approach in which the spectral irradiance is divided into two subsets, and each subset is modeled with a different network. The results show that the RBF neural network model produces well generalizations while the Planck-polynomial model produces poor ones. During the generalizations, the maximum relative deviation of the RBF neural network model and the Planck-polynomial model were 0.027% and 3.46%, respectively.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Binhua Chen, Caihong Dai, Zhifeng Wu, Lei Fu, "Curve fitting for standard lamp of spectral irradiance based on RBFNN", Proc. SPIE 9046, 2013 International Conference on Optical Instruments and Technology: Optoelectronic Measurement Technology and Systems, 904613 (19 December 2013); doi: 10.1117/12.2037491; https://doi.org/10.1117/12.2037491
PROCEEDINGS
8 PAGES


SHARE
Back to Top