Monitoring the Rice Panicle Blast Control Period Based on UAV Multispectral Remote Sensing and Machine Learning

The heading stage of rice is a critical period for disease control, such as panicle blast. rapid and accurate monitoring growth great significance plant protection operations in large areas mobilizing resources. For this paper, the canopy multispectral information acquired continuously by an unmanned aerial vehicle (UAV) was used to obtain rate inversion. results indicated that multi-vegetation index inversion model more than single-band single-vegetation models. Compared with traditional algorithms neural network (NN) support vector regression (SVR), adaptive boosting algorithm based on ensemble learning has higher accuracy, correlation coefficient (R2) 0.94 root mean square error (RMSE) 0.12 model. study suggests effective UAV remote sensing can be built using AdaBoost index, which provides crop acquisition processing method determining timing tassel control.