A method for extracting features of large-area terrain changes based on unmanned aerial vehicle surveying and mapping images
In order to solve the problem of poor feature extraction performance for large-scale terrain changes,a feature extraction method for large-scale terrain changes based on unmanned aerial vehicle(UAV)surveying and mapping images was proposed.Image information captured by UAVs was collected using the information reconstruction method,and the terrain edge scale was calculated.The large-area images were preprocessed using image filtering algorithms.Terrain feature parameters were measured and calculated,and terrain change features were decomposed using weighted full polarization.The feature differentiation effect was enhanced through feature selection factors and feature judgment factors,and terrain features were classified using synthetic aperture radar(SAR)feature classification methods.The optimal weighted SAR feature dataset was processed through random forests and classifiers.Terrain change feature extraction was completed based on the voting results of the decision tree.Experiments have shown that after the application of the proposed method,the size of large-scale terrain features best matches the actual values,and the time required to extract features from mountainous and flat areas is shorter.The overall feature extraction effect is good,making it easy to quickly adjust the surveying and mapping path and parameters and thus achieving real-time monitoring of terrain changes.
国家科技期刊平台
NSTL
万方数据
