An Intelligent 3D Mapping Method for Buildings Based on Multi-source Remote Sensing Data

Abstract

The invention discloses an intelligent three-dimensional mapping method for buildings based on multi-source remote sensing data, which uses the aerial remote sensing image data and laser point cloud data of the target survey area to obtain the building roof vector profile of the target survey area through the following steps , Structured 3D wireframe model of buildings. The present invention uses aerial remote sensing image data to generate image three-dimensional point cloud, and completes high-precision registration and fusion of image point cloud and laser point cloud; uses the fused three-dimensional point cloud to generate high-precision DSM, DEM, traditional DOM and OESM models; uses three-dimensional The point cloud extracts the building boundary, extracts the contour line on the DOM, and combines the two to extract and filter the two-dimensional vector contour of the building to achieve high-precision extraction of the building roof contour; based on the two-dimensional vector contour of the building roof and high-precision OESM Model quickly generates a structured wireframe 3D model of a building. The invention does not need manual intervention, and realizes fully automatic three-dimensional mapping of buildings.

Description

technical field [0001] The invention belongs to the field of three-dimensional mapping of buildings, in particular to an intelligent three-dimensional mapping method for buildings based on multi-source remote sensing data. Background technique [0002] The production efficiency of the traditional stereo mapping method is low, and it takes a lot of labor and time to realize the semi-automatic extraction of building outline information by manual intervention. Most of the existing 3D modeling methods for large-scale buildings are based on large-scale point cloud information for network reconstruction, which has high computational complexity, and the generated models do not have structured information, and it is desired to obtain a single 3D model of the building raletively hard. [0003] The Chinese patent with the publication number CN109685891A applied for a 3D building modeling and virtual scene generation system based on depth images. The patent implementation method inclu...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that manual interactive editing is required to achieve precise line feature constraints, and there are more requirements for data sources

Images