Assessment Of The Gully Morphology Measurement Method Based On UAV Photogrammetry

Gully erosion is an intense type of soil erosion with three-dimensional morphological characteristics in space,and once formed,gully erosion can cause severe land degradation and environmental degradation.The development of unmanned aerial vehicles(UAVs)has made it possible to quickly and easily apply UAV photogrammetry to the study of gully erosion.There is a lack of uniform solutions for efficient acquisition and processing of data in UAV measurement studies of gully erosion patterns,and few studies have been conducted on the accuracy and error of topographic data.The purpose of this study is to evaluate the accuracy of data by analyzing the errors of different schemes,to analyze the processing efficiency of data,and to provide suitable scheme design ideas for UAV’s gully erosion measurement research.Yuanmou Dry-hot Valley gully is selected as the research object,different flight schemes and ground control point(GCP)schemes are used to acquire and process the data,and finally the errors are analyzed by Digital orthophoto model(DOM)and digital surface model(DSM)to explore the environmental interference factors affecting the measurement accuracy and optimize the UAV survey scheme for the gully morphology.The conclusions of the paper are as follows:(1)Due to the significant differences in the elevation and morphology of the gully channel on gully shoulder line,gully wall,gully bed and other parts,the number and location of GCP also have a significant impact on the accuracy,and the introduction of 3 reasonable deployment of GCP in the data processing can reduce the error by 24.5 %.It was found that the number of 9 GCPs and comprehensive consideration of its level and elevation of the uniform distribution scheme is the best,which was more efficient and could improve the accuracy by 22.7 %.When the flight altitude is low,the number of ground control points can be appropriately reduced under the premise of reasonable layout scheme.(2)Among all the schemes,the scheme with a flight height of 30 m,80% /70 %photo overlap and 11 ground control points has the highest accuracy(mean absolute error of 0.0353 m and root mean square error of 0.0525 m),but this takes more time for data collection and processing and is less efficient.The altitude has a large impact on the accuracy of gully topography data,and the accuracy of gully bed is the lowest at the same flight altitude,and the error of gully bed changes most obviously when the flight altitude changes.In the flight altitude of 100 m,the influence of overlap on monitoring accuracy is relatively not significant,and the MAE and RMSEz of all schemes fluctuate around 7 cm and 13 cm,respectively,and large errors occur at the gully shoulder line in some schemes.(3)When the flight altitude is about 65 m,the photo overlap is not less than 50%/ 40%,the number of control points is 9 and the layout is reasonable,the DSM average elevation accuracy can reach about 4 cm,the monitoring results can simultaneously take into account both accuracy and measurement efficiency.The sources of errors and the distribution locations of check points with large errors were analyzed from four aspects: shadow,slope,slope direction and vegetation,and it was found that different light conditions due to the morphology of the gully and vegetation would lead to some differences in the errors of the DSM elevation.In the range of slope less than 50°,the correlation between error and slope is small.The vegetation growing in the gully introduces errors in DSM and its effect on the accuracy is difficult to eliminate.The use of UAVs in gully erosion studies is very convenient and fast,and can yield post-products with centimeter-level accuracy.The paper uses UAV photogrammetry to investigate and study the morphology of gully erosion in the Yuanmou Dry-hot Valley area,obtains gully data through different ground control point schemes and flight schemes,analyzes and discusses the accuracy,data volume and data processing efficiency of the later digital products,identifies error sources,and provides program design ideas and research method support for efficient,fast and convenient acquisition of gully morphology data in the Dry-hot Valley.