Research On Regional Detection And Analysis Method Of Bursaphelenchus Xylophilus Based On Aerial Remote Sensing Data

For pine trees,the damage of Bursaphelenchus xylophilus is fatal,and even in the worst case,the original ecological balance of the forest system will be destroyed.Pine trees account for 25% of China's forest resources,and its health and forestry development are closely related.Pine wood nematode has a high speed of transmission and a high mortality rate,which has great difficulty in prevention and treatment.Nowadays,experts and scholars at home and abroad have not given a good solution to the identification and detection of pine wood nematode disease.Although the identification of pests and diseases from aerial video and satellite remote sensing images can achieve good results,there are still problems of high cost and insufficient accuracy,which leads to the unsatisfactory effect of popularization and application in the field of agricultural and forestry pest detection.As the development of helicopter technology has matured,its performance has been continuously improved and improved,and it has occupied a considerable share in the military and civilian fields.The helicopter is used as a remote sensing platform,and the multi-spectral sensor is used to collect remote sensing data.It plays an important role in forest fire prevention,land use survey,coal mine resource survey and environmental protection.In particular,the continuous outbreak of natural disasters in China in recent years has led to corresponding developments in the research of agricultural and forestry pest and disease monitoring.However,the aerial remote sensing image data is large,and the ground object information is complex.How to quickly analyze and accurately identify the pest and disease information from such a large and complex remote sensing image is the most important subject of the current pine wood nematode identification and detection research.In this paper,a multi-spectral camera is mounted on a helicopter to obtain visible and near-infrared remote sensing images of the forest area to be detected.The professional remote sensing software ENVI is used to analyze the image to find out the characteristics of the dominant lethal tree,and then combined with image classification processing to achieve the identification and detection of trees infected by pine wood nematode disease,so as to find out the actual location of the lethal tree to facilitate manual intervention.The main research contents of the thesis are as follows:1.The purpose and significance of the identification and detection of pine wood nematode disease were expounded.The research status at home and abroad was introduced.The theoretical basis for identifying pests and diseases in remote sensing images was introduced.The remote sensing image preprocessing method and the commonly used algorithms in remote sensing image classification process were introduced.Spectral,texture,and spatial features in the image were analyzed.2.Based on the disadvantages of spectral classification pixels,based on this,a rule-based object-oriented classification method is proposed to extract pests and diseases,combined with the corresponding attributes of the three rules,and the misclassification of spectral classification is eliminated.The experimental results show that the method accurately classifies the diseased trees on the remote sensing image.Compared with the manual detection,the method can save the time of the analysis and achieve higher precision requirements,which meets the needs of practical applications.3.Based on the accurate classification of disease trees,the system is implemented by using C# combined with GDAL components,marking the location of the disease tree on the remote sensing image,obtaining the real geographical location of the diseased tree according to the location of the marker,and then conducting manual intervention.