Quantification of Physiological Parameters of Rice Varieties Based on Multi-Spectral Remote Sensing and Machine Learning Models

Estimating plant physiological indicators with remote sensing technology is critical for ensuring precise field management. Compared other platforms, low-altitude unmanned aerial vehicles (UAVs) produce images high spatial resolution that can be used to clearly identify vegetation. However, the information of UAV image data relatively complex and difficult analyze, which main problem limiting its large-scale use at present. In order monitor indexes from multi-spectral data, a new method based on machine learning studied in this paper. Using deriving absorption coefficients canopies whole leaf area, paper quantifies effects such as soil analyzer development (SPAD) value, dry matter accumulation relationship between reflectance spectra. Nine vegetation were then extracted sensitive rice indicators. SVM model predict SPAD value plant, mean squared error (MSE), root (RMSE), absolute (MAE), percentage (MAPE), symmetric (SMAPE) values 1.90, 1.38, 0.13, 0.86, 4.13, respectively. The results demonstrate plants display considerable biochemical spectral correlation. has better effect because adaptation higher accuracy than models. This study suggests acquired using quickly estimate indicators, potential pre-visual detection field. At same time, it also extended inversion key variables crops.