14 October 2015 Estimation of leaf chlorophyll content in winter wheat using variable importance for projection (VIP) with hyperspectral data
Author Affiliations +
Abstract
Accurate estimation of leaf chlorophyll content (LCC) has great significance in study of the winter wheat, which is important for indicating nutrition status and photosynthetic. Selecting the closed related variable is the key to LCC monitoring. The variable importance for projection (VIP), applied to little samples and strong correlation data, is one of variable selection methods. In this study, VIP was used to select spectral variables, which includes reflectance spectra, first derivative spectra, vegetation indices and absorption or reflectance position features. The grey relational analysis (GRA) was used as a comparison. The results showed that (1) the VIP technology could be used to variable selection and had a strong correlation. (2) Reflectance spectra with the VIP method displayed the best accuracy, with R2 and RMSE of 0.42 and 0.663mg/g, respectively. (3) Vegetation indices using GRA had higher estimation than VIP method, with R2 and RMSE of 0.52 and 0.607 mg/g, respectively. (4) The VIP had more superiority and higher accuracy than the GRA in all kinds of hyperspectral features except vegetation indices. Therefore, the VIP technology could be used to the estimation of LCC and had a relatively good accuracy.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Peng He, Peng He, Xingang Xu, Xingang Xu, Baolei Zhang, Baolei Zhang, Zhenhai Li, Zhenhai Li, Haikuan Feng, Haikuan Feng, Guijun Yang, Guijun Yang, Yongfeng Zhang, Yongfeng Zhang, } "Estimation of leaf chlorophyll content in winter wheat using variable importance for projection (VIP) with hyperspectral data", Proc. SPIE 9637, Remote Sensing for Agriculture, Ecosystems, and Hydrology XVII, 963708 (14 October 2015); doi: 10.1117/12.2195465; https://doi.org/10.1117/12.2195465
PROCEEDINGS
12 PAGES


SHARE
Back to Top