Rice blast is one of the major rice diseases recognized in the world, and it is also the most serious rice disease in Northeast China. This study is based on the natural disease in the field, using a six-rotor drone equipped with a six-channel multi-spectral camera Micro-MCA6 Snap to obtain multi-spectral data of the diseased field. According to the results of the canopy scale research and applying the variance analysis method, 19 characteristic spectra are selected. The regional scale rice panicle blast classification models were constructed by three classification methods of stepwise linear discriminant analysis, support vector machine and neural network respectively. The experimental results showed that the recognition accuracy of various models was above 87.5%. The best classification effect was the discriminant analysis model. It only selected 3 features to achieve a cross-validation accuracy of 90.6% and a predicted sample recognition accuracy of 92.8%.
Research on Recognition of Rice Panicle Blast in Cold Region Based on UAV
Lect. Notes Electrical Eng.
International Workshop of Advanced Manufacturing and Automation ; 2020 ; Zhanjiang, China October 12, 2020 - October 13, 2020
2021-01-23
7 pages
Article/Chapter (Book)
Electronic Resource
English
UAV remote sensing , Rice panicle blast , Spectral characteristics , Fisher discriminant analysis , Neural network , Support vector machine Engineering , Control, Robotics, Mechatronics , Manufacturing, Machines, Tools, Processes , Engineering Economics, Organization, Logistics, Marketing , Artificial Intelligence
Research on Recognition of Rice Panicle Blast in Cold Region Based on UAV
British Library Conference Proceedings | 2021
|A Lightweight YOLO Model for Rice Panicle Detection in Fields Based on UAV Aerial Images
DOAJ | 2024
|Transcriptomic analyses of space-induced rice mutants with enhanced susceptibility to rice blast
Online Contents | 2007
|Method and model for predicting chlorophyll content of rice canopy in cold region
European Patent Office | 2022
|