Research on Remote Sensing Monitoring of Wheat Scab and Stripe Rust Based on GF-6 Satellite Data
He Fuwei
Tian Hongwei
Zhao Zhiyu
Wang Haiyan
Abstract：Scab and stripe rust are common diseases in the main wheat planting area in Henan Prov-ince.The dynamic monitoring of scab and stripe rust is of great significance for disease prevention and control as well as high and stable yields of wheat.To investigate the application of China's high-resolu-tion satellites in the regional monitoring of scab and stripe rust,this study developed multiple vegetation indices based on GF-6 satellite data.Moreover,combined with meteorological factors including air tem-perature and relative humidity in March and April,remote sensing monitoring models for scab and stripe rust in winter wheat were established by Multiple Linear Regression(MLR)and Back Propagation Neural Network(BPNN),respectively.Additionally,monitoring research was conducted on the disease inci-dence rate of spikes(scab)and infected leaf rate(stripe rust)in winter wheat planting areas of southern Henan Province.The results showed that the incorporation of meteorological factors significantly enhanced monitoring accuracy for the disease incidence rate of spikes in scab.The coefficient of determination(R2)of the MLR and BPNN models increased from 0.42 and 0.39 to 0.67 and 0.82,respectively,while their root mean square errors(RMSEs)decreased from 3.99 and 4.16 to 3.03 and 2.21,respec-tively.The BPNN model exhibited superior performance compared to the MLR model.For the infected leaf rate in stripe rust,the integration of meteorological factors improved the monitoring accuracy of the MLR model,with R2 rising from 0.19 to 0.26 and RMSE decreasing from 18.56 to 17.79.However,no significant improvement was observed in the BPNN model.
Keywords：GF-6 satellitewinter wheatscabstripe rustBP neural network
Publication Date：2025-05-30
Online Publishing Date：2026-09-12(First online date of this platform, not the publication date of the document)
Pages：9( 94-102 )