Research On Spatial Information Extraction Technology Of Rural Settlements Based On UAV Low-altitude Remote Sensing Images

In recent years,with the rapid development of UAV technology,computer technology,geographic information data can be acquired is more abundant than single satellite remote sensing data,the data processing ability has been improved greatly,so the UAV remote sensing technology has been used widely in recent years.Although the satellite remote sensing technology,aircraft based in most areas of China is still the main means of remote sensing satellite,but for the Southwest Mountainous foggy area is usually difficult to obtain high-resolution image data,and the use UAV can fly in these areas,so as to obtain the high resolution image data,which can also identify the type of attribute information of object,to better achieve the extraction of spatial information on rural land.The traditional technology of universal remote sensing image classification accuracy is not high,it is difficult to achieve effective extraction of UAV high resolution image information,has a very important significance to adopt a new method to improve the extraction technology of UAV high resolution image of rural settlement spatial information accuracy.In this paper,we have discussed the process of acquiring image data and processing data by UAV remote sensing technology,and obtained clear and good quality images.This paper selects some rural settlements in Luoping County of Qujing City,Yunnan Province as the research area,and introduces the composition,aero photography process and data processing methods of UAV remote sensing system.According to the process of route design,aerial photography,image preprocessing,image matching,connection point and space three encryption,we get the high-resolution UAV remote sensing image of the study area,and finally generate the orthophoto map of the survey area.The object oriented method is used to segment the remote sensing image.The best segmentation threshold,merging threshold and texture convolution kernel parameters are obtained according to the information of rural settlements.The segmentation results under different threshold conditions are compared.Technology segmentation by object oriented classification,establishes corresponding to different objects,According to the characteristic information of house,woodland,water surface,bare land and farmland,spectral information and spatial structure of the study area were automatically extracted by standard nearest neighbor method,and the accuracyof the extraction results was evaluated through the confusion matrix.The paper summarizes the whole process from the acquisition and processing of UAV data to the extraction of rural settlement spatial information,and induces the classification rules of rural settlement spatial information.The research of this paper provides a certain guiding significance for the use of unmanned aerial vehicle(UAV)high resolution remote sensing image to extract land use information.