Research On 3D Change Detection Of Urban Construction Accumulation Based On Multi-source Remote Sensing Data

In recent years,due to economic development and the increasing demand of the people for urbanization,construction-related industries have developed rapidly,and various engineering constructions have also led to a continuous increase in the production of construction accumulation.Real-time monitoring and precise control of construction accumulation accumulations seems very important.At present,most of the change detection for construction accumulation accumulations focuses on two-dimensional spatial changes,ignoring the impact of changes in elevation space,and most of the volume calculations for construction accumulation accumulations are for a single accumulation,which is cumbersome and less efficient,which is safe for construction accumulation.There is not enough attention to the risk of instability.Taking into account the complexity of the three-dimensional changes of the construction accumulation,this article uses the remote sensing image of Resources satellite three,Phantom 4 RTK aerial survey image and Focus3 D point cloud data,and uses Pingdingshan City,Henan Province as the research area.This kind of data is used to detect the three-dimensional changes of construction accumulation accumulations at different scales.The research content mainly includes the following aspects:(1)Using the three-dimensional images taken by the multi-line array camera of Resources satellite three,the construction accumulation and change detection are carried out on the scale of the local area of the city.According to the multi-view characteristics of Resources satellite three,based on the production of the research area DSM,the classification method based on elevation features is used to extract construction accumulation and buildings,and the adaptive roughness is used to further extract the construction accumulation area,through the threedimensional difference method Obtain the DSM elevation difference images of different time phases,and determine the elevation change range of the construction accumulation through the adaptive threshold method.The experimental data proves that the extraction accuracy of construction accumulation can reach 77.57% in area.(2)Based on aerial drone photogrammetry data,the accumulation body is extracted and three-dimensional change detection on the scale of the construction accumulation.Based on the generation of DSM and DOM in the experimental area,combined with the method of supervised classification and adaptive roughness,the construction accumulation is extracted,and the parameter-based discrete integral method is used to calculate the volume,slope,aspect and other parameters of the construction accumulation.Analyze the construction accumulation with a large degree of change,analyze the main reasons of its three-dimensional change,and explain the main impact of its change.Experimental verification shows that the extraction accuracy of construction accumulation can reach 80.6%.(3)Use air-ground fusion to obtain point cloud data of a single construction accumulation,use principal component analysis and K-D tree-based ICP algorithm to fuse the data,and verify the accuracy of the data.Using the analytic hierarchy process to analyze the instability of the construction accumulation on the UAV photogrammetry data and the ground point cloud data.The analysis results show that the instability risk prediction is instructive.The experimental results show that the use of satellite remote sensing,aerial drones and ground 3D laser scanning,combined with the classification method of elevation features,as well as methods such as supervised classification and adaptive roughness,can achieve the extraction of the 3D features of the construction accumulation.Among them,this paper introduces the ground roughness to the extraction of construction accumulation,and combines it with the roughness coefficient.The slope length is used as a parameter to form the adaptive roughness.The experimental results show that on the multi-source remote sensing data,the adaptive roughness can be used in the elevation texture.There is a targeted extraction of construction accumulation.The detection results of the three-dimensional change of the construction accumulation show that the use of parameter-based discrete integral method to calculate the volume of the construction accumulation and other three-dimensional parameters can effectively obtain the volume information of the construction accumulation,and the analytic hierarchy process is used to lose the construction accumulation.Stability research to realize the risk prediction of the instability of the accumulation body.On the basis of obtaining multi-source data,the principal component analysis method and the KD tree-based ICP algorithm are used to fuse the data.The results meet the model accuracy requirements of the relevant specifications.The experimental results prove that the fusion data can more accurately and completely represent the construction accumulation Three-dimensional model of the stacked body.It can be proved that the use of multi-source remote sensing data to observe the construction accumulation can realize the three-dimensional change detection of the spatial distribution,elevation,volume and other characteristics of the construction accumulation,and carry out the loss of the construction accumulation according to the three-dimensional model.Stability analysis,monitoring the risk of instability of construction accumulation accumulations.Provide technical support and data foundation for the precise management and control of construction accumulation.