ASABE Technical Library - Abstract
Member and Access Notice
Prediction of Chlorophyll Content of Winter Wheat using Leaf-level Hyperspectral Imaging Data
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2009 Reno, Nevada, June 21 - June 24, 2009 096434.(doi:10.13031/2013.27133)Authors: Yankun Peng, Wei Wang, Hui Huang, Xiu Wang, Xiaodong Gao
Keywords: Winter wheat; Chlorophyll content; Leaf-level hyperspectral imaging; Partial least-squares regression (PLSR); Vegetation index.
The leaf-level winter wheat hyperspectral response to its chlorophyll content was examined in this study. And the main objectives was to assess the possibility of estimating biochemical chlorophyll a and chlorophyll b of field winter wheat using hyperspectral images and to estimate the obtained model accuracy. Firstly, the cube image data was constructed and the region of interest (ROI)was selected, then after the average pixel intensity acquired, using stepwise discrimination method the optimal wavelengths were selected; then the chlorophyll content model using partial least-squares regression (PLSR) was constructed and validated; Secondly, we computed three representative vegetation indices, such as normalized difference vegetation index (NDVI), chlorophyll absorption reflective ratio index (CARI) and pigment specific normalized difference (PSND), and compared their ability to predict chlorophyll content with the PLSR method. Results demonstrated that using the model constructed using PLSR based on the 12 optimal wavelengths, we can get a fairly good result, that is R2=0.9216 and RMSE=0.2105. And the result of this method was better than those of vegetation indices.
(Download PDF) (Export to EndNotes)