Research On Corn Diseases And Pests In Jilin Province Based On Meteorological And Remote Sensing Data

Jilin Province is an important corn production area in my country.The annual average corn output and export volume are in the forefront of the country.Therefore,ensuring the yield and quality safety of corn crops has important practical significance for the economic development and food security of Jilin Province.However,in the growth process of corn,it is facing a major threat of diseases and insect pests.According to statistics,the annual reduction of corn production can range from 20%to 30% due to diseases and insect pests in Jilin Province,and the harvest is severe.Therefore,understanding the main types of diseases and insect pests currently faced by corn and summarizing their occurrence rules,and formulating scientific and reasonable control measures on this basis are very important for ensuring the healthy growth of corn crops and improving its economic benefits.The traditional means of identification and monitoring of diseases and insect pests is mainly through manual field survey.This method is not only inefficient,but only suitable for monitoring on a small area scale.It is difficult to carry out in the modern large-scale corn planting,and it has a lag in the forecasting of disease and insect information.It affects the time and effect of prevention and treatment.The emergence and vigorous rise of remote sensing technology provides a new way for corn disease and insect pest monitoring and early warning.Remote sensing has the characteristics of fast,accurate,real-time,and large area data can be obtained in a short period of time.Remote sensing can monitor the growth of crops without direct contact with the crops,so it has become an excellent means of monitoring crop diseases and insect pests.Using ground-measured spectral data and sampling point data,GF-1 WFV images,Landsat8 images,PL images,drone data,and weather data from 2019 to 2020 are used for corn disease and insect pest monitoring and early warning.This article first analyzes the changes in spectral reflectance of corn under the stress of red spider and big leaf spot disease,and discusses the mechanism of effect of red spider and big leaf spot disease on corn crops.It is found that after corn is invaded by diseases and insect pests,the spectrum in the visible light region and the near-infrared region will be obvious.Changes,among them,the near-infrared zone changes are particularly obvious,indicating that the near-infrared zone has more application value in actual corn change monitoring.In addition,the change of reflectivity of maize in milk maturity period is more significant than that in maturity period,indicating that milk maturity period is more suitable for monitoring corn diseases and insect pests than mature period.In addition,this study used ground-measured spectral data to extract the vegetation index,analyzed the correlation between the vegetation index and the severity of corn affected by red spider and large leaf spot disease,and screened the optimal vegetation index EVI for identifying the severity of corn affected by diseases and insect pests.,Use the support vector machine method to extract the maize planting area in the study area,and get the remote sensing image maize EVI through the mask.A threshold standardization method was established based on EVI,and a general monitoring model for corn red spider and large leaf spot disease was established using decision tree classification and natural discontinuity classification methods.Field sampling points were used to verify the accuracy of the model.The results showed that the large-scale corn disease and insect pest monitoring Above,the accuracy of the decision tree classification model is better than the natural discontinuity classification model,with an accuracy of up to 66.7%,and the monitoring model is still applicable to the damage monitoring of corn armyworm in Jilin Province and the corn red spider in Kailu County,Inner Mongolia in 2018.The monitoring accuracy at the town level can reach up to 83.33%,indicating that the model has good universal applicability.In addition,the use of UAV data to verify the plot-level monitoring accuracy found that the UAV data monitoring results are generally consistent with the GF-1 WFV satellite image monitoring results,which once again proves that the model has good accuracy and can provide technology for actual production.stand by.In the early warning of corn red spider,big leaf spot and armyworm,this paper combines remote sensing data and meteorological data,and uses the maximum entropy niche model method to warn corn red spider,big leaf spot and corn armyworm.At the same time,it uses The multiple linear regression method provides early warning of the occurrence risk of armyworm.The results show that the maximum entropy niche model has a better early warning effect and is better than the multiple linear regression model.The early warning results highly overlap with the actual occurrence of pests and diseases,and the warning accuracy can reach more than 80%.It shows that based on the principle of maximum entropy,combined with remote sensing and meteorological data,it is feasible to warn the occurrence risk of corn red spider,large leaf spot and armyworm.