Hybridization of partial least squares and neural network models for quantifying lunar surface minerals

Article ID	Journal	Published Year	Pages	File Type
10701689	Icarus	2012	18 Pages	PDF

Abstract

âº A hybridized PLS-BPNN model is developed for quantifying lunar surface minerals. âº The model is validated with the LSCC dataset and additional 12 Apollo samples. âº The model can assess minerals and chemistry for Apollo soils non-destructively. âº The model is suitable for hyper- and multi-spectral data to quantify minerals. âº We try to understand the physical meaning of BPNN by analyzing loading curves.

Keywords

Data reduction techniques Regoliths Moon Moon, surface

Related Topics

Physical Sciences and Engineering Earth and Planetary Sciences Space and Planetary Science

Preview

Hybridization of partial least squares and neural network models for quantifying lunar surface minerals

Authors

Shuai Li, Lin Li, Ralph Milliken, Kaishan Song,