作者: Wenqian Huang , Yan'an Wang , Jianhua Guo , Zhiming Guo , Chunjiang Zhao
DOI: 10.1080/00387010.2013.816748
关键词: Chemistry 、 Least squares 、 Calibration (statistics) 、 Partial least squares regression 、 Biological system 、 Spectroscopy 、 Chlorophyll 、 Analytical chemistry 、 Interval (mathematics) 、 Mean squared error 、 Correlation coefficient
摘要: ABSTRACT Chlorophylls respond rapidly to the current physiological status of a tree and reflect nutrient availability. Visible/near-infrared spectroscopy was attempted determine foliar chlorophyll content in an apple orchard. Backward interval partial least squares genetic algorithms were sequentially applied select optimized spectral combination regions selected from informative model calibration. used remove noninformative regions, which significantly reduced number variables. The subsequent application algorithms-partial this domain could lead efficient refined model. performance final back-evaluated according root mean square error calibration (RMSEC) correlation coefficient (R c ) set, then tested by prediction (RMSEP) ...