Estimation of grain protein content in winter wheat by using three methods with hyperspectral data
2016
Xiu-liang, J. (Beijing Research Center for Agri-Food Testing and Farmland Monitoring, Beijing (China)) | Xin-gang, X. (Beijing Research Center for Information Technology in Agriculture, Beijing (China)) | Hai-kuan, F. (National Engineering Research Center for Information Technology in Agriculture, Beijing (China)) | Xiao-yu, S. (National Engineering Research Center for Information Technology in Agriculture, Beijing (China)) | Wang, Q. (National Engineering Research Center for Information Technology in Agriculture, Beijing (China)) | Ji-hua, W. (Yangzhou Univ., Yangzhou (China). Key Lab. of Crop Genetics and Physiology of Jiangsu Province) | Wen-shan, G.
Grain protein content (GPC) is an important quality indicator for cereal crops to meet variety of needs of commodity. The objectives of this study were: 1) to analyze relationships between the single vegetation indexes and GPC; 2) to improve estimation accuracy of GPC by using two or three vegetation indexes and partial least squares method (PLS); 3) to compare the performance of the proposed models by the three methods. Vegetation indexes and concurrent GPC of samples were selected in Xiaotangshan experimental sites, Beijing, China, during 2008-2009, 2009-2010 and 2011-2012 winter wheat growth seasons. This study showed that the GPC could be effectively estimated using three methods (single vegetation indexes, two or three vegetation indexes and PLS). The lowest RMSE and highest R2 were PLS(b) regression model for PLS (R2=0.63 and RMSE=0.615%); SIPI, OSAVI and MTVI2 for three vegetation indexes (R2=0.57 and RMSE=0.84%); MTVI2 and MTCI for two vegetation indexes (R2=0.56 and RMSE=0.92%) and DCNI I for single vegetation indexes (R2=0.53 and RMSE=1.12%), respectively. The PLS(b) was better than others methods for estimating GPC in winter wheat. But two or three vegetation indexes also has its own merit, particularly when taking into consideration the simplicity of its application. This method may be provide guideline for improving the estimation accuracy of winter wheat GPC in a large worldwide by using different algorithm and satellite images data.
显示更多 [+] 显示较少 [-]