Research On Fine Mapping Of Power Corridor Based On UAV Remote Sensing

With the improvement of economy and comprehensive national strength,China’s power grid construction scale has ranked into the top of the world.At present,most of the transmission line construction data is mainly obtained by the traditional methods such as satellite remote sensing and conventional aerial photogrammetry technology.Compared with the traditional methods,UAV remote sensing has the advantages of easy data acquisition,high accuracy and high efficiency.Therefore,it is of great research significance to carry out fine mapping research on power corridor based on UAV remote sensing.Based on this point,this paper takes Xiamen Jimei District as the research area,based on the UAV remote sensing technology,carrying out three aspects of cartographic methods,there are the topographic map of power corridor,the power line net height analysis map and the power disaster map.Finally,a set of high-precision and rapid power corridor fine drawing method for power corridors were obtained,which laid a theoretical and practical foundation for actual power planning and design,inspection and post-disaster emergency.The main research contents are as follows:(1)Large-scale topographic map production in power corridor planning.Compared with the traditional large-scale topographic map production,in this paper,a novel 3D model of UAV tilt photogrammetry is introduced to make line drawing,the cluster calculation and 3D model refinement are introduced to improve the accuracy and efficiency of drawing,as well as used the built-in deep learning module of ArcGIS Pro to statistically classifies and maps the number of plants under the power corridor.The verification results show that the method based on uav image and digital mapping can meet the requirements of 1:500 mapping accuracy.The results show that this method can not only meet the precision requirements of 1:500 large-scale topographic map production,but also greatly improve the efficiency of mapping and reduce the workload of the field.It provide a good data basis for power planning design and inspection.(2)The power line net height analysis diagram in power corridor operation and maintenance.Based on the UAV image,this paper proposed a binocular stereo measurement method to extract the height of the transmission line,and used the laser radar method to measure the height of the power line and trees to verify the accuracy.The results show that the median error between the power line extracted by stereo measurement and the height of the tree compared with that measured by lidar method is 0.21m and 0.12m,and the extraction accuracy up to submeter level.(3)Quick drawing of power disasters during emergency power protection.Based on the orthophoto image output from the remote sensing data of the UAV,the Alex Net deep convolution neural network is used to extract the lodging pole after typhoon and verify its accuracy through field investigation,and the accuracy was verified through field investigation.The results show that the extraction accuracy of this method reaches 71%.It can be seen that the accuracy of the power disaster map produced by this method is high,which can quickly provide intuitive and comprehensive first-hand information for post-disaster power emergency rescue and coordinated decision-making.