Multi Factor Synergistic Estimation Of Winter Wheat Biomass And Scab By Rmote Sensing

As a typical climate disease,The occurrence of Fusarium head blight of winter wheat is more and more frequent,which seriously affects the yield and quality of Winter Wheat in China.It is important to predict the epidemic degree of winter wheat scab at regional scale in advance.Remote sensing technology is widely used in crop growth monitoring and crop diseases and insect pests monitoring and prediction because of its timely access to continuous spatial information and large monitoring range.It is far from being able to meet the requirements of today’s agriculture by relying on one technical means.The problem of forecasting diseases and pests in time is gradually highlighted,Which is established by combing with meteorological data and remote sensing data.This pest prediction model is simple,high precision and high universality.This study build multiple linear regression model and BP neural network model,which winter wheat scab was studied as an object of study.These model based on the survey data of field winter wheat scab on the county level,and combined with meteorological data and multiple source remote sensing data.Then analyzing and comparing the advantages and disadvantages of the two model,and finally realizing the grade prediction of Winter Wheat Scab in Dafeng District,and analyzing the occurrence of scab.The results of the study are summarized as follows:(1)The biomass of winter wheat at jointing stage was estimated by modified winter wheat biomass model.The characteristics and changes of biomass distribution in the heading stage of winter wheat were further analyzed based on HJ satellite remote sensing image data.The determining coefficient of biomass and measured values of winter wheat biomass model was 0.9191,and RESE was 214.8 kg hm-2.Winter wheat biomass model has high estimation accuracy;The biomass of winter wheat heading stage than jointing stage biomass changed significantly,the growth of fast change of farmland area is 20108.7 Hm2,accounting for 23.4%of the total planting area,mainly distributed in the northeast of Shuyang county.(2)The index of winter wheat scab index is related to NDVI,relative humidity and biomass,and has a linear correlation with LAI,which is related to the power exponent function of the temperature.The study is divided into training set and test set,and the multiple linear regression model of winter wheat scab is constructed by SPSS.The regression model equation is Y=2.801X1+0.001X2+0.15X3+0.419X4+3.248X5.The results showed that the scab index of Dafeng area was close to the actual measured index of winter wheat scab disease,which was concentrated on both sides of 1:1 line.Meanwhile,the RMSE value of the model was 2.2612,and the fitting result of the model was ideal.(3)According to the multiple linear regression model,the BP neural network algorithm was used to establish the estimation model of winter wheat scab.The model R2 is 0.8244,and the RMSE value is 2.3778.Comparing the advantages and disadvantages of the two models.The accuracy of the two models is quite different.But the degree of discretization between the estimated value and the measured value of the BP neural network model is inferior to the regression model.Finally,the multiple linear regression model was used to estimate the winter wheat scab in Dafeng district.The results are visualized on the space map.The occurrence of scab of Winter Wheat in Dafeng district was moderate and light.The scope of the disaster prediction is wide,but lighter.The heavily region was distributed in the South and southwest,especially Dongbatou farm area.It is suggested that using HJ star data to predict the severity of winter wheat scab is feasible,and has a certain reference value for the prediction and control of wheat scab.