3D Building Segmentation from Aerial LIDAR Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for automatic 3D building segmentation using aerial LIDAR data are inefficient and inaccurate, particularly in dense urban areas, as they assume ground covers the largest surface area, which is inappropriate for such regions.
Innovation Solution
A method involving generating a point cloud from LIDAR data, classifying points into ground and non-ground, segmenting non-ground points into buildings and clutter, and calculating a fit for rectilinear structures with the fewest vertices, using techniques like ball pivoting, expectation maximization, and loopy belief propagation to ensure accurate segmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional ground-based data collection is used, then high detail on building facades is achieved, but information on building tops is lost
Solution Approach 1:
The patent combines ground-based LIDAR data collection with aerial LIDAR data to create a comprehensive 3D building model. The ground-based system captures detailed facade information while the aerial system captures rooftop information, and both datasets are merged into a single integrated model, resolving the contradiction between facade detail and rooftop coverage.
2Area of stationary object
If aerial data is used, then accurate building footprints and large area coverage are achieved, but building side information is lost
Solution Approach 1:
The patent merges aerial LIDAR data, which provides accurate building footprints and large area coverage, with ground-based LIDAR data, which captures building side and facade information. This combination allows the system to maintain both wide coverage and detailed side information that would be lost using aerial data alone.
3Ease of manufacture
If 2D imagery is used for building modeling, then low acquisition cost is achieved, but automatic 3D modeling accuracy deteriorates
Solution Approach 1:
The patent replaces 2D image-based modeling with direct 3D LIDAR sensing. LIDAR actively measures distance and depth by emitting laser pulses and measuring their return time, providing direct 3D point cloud data without relying on stereo algorithms or photogrammetry, thereby achieving high 3D modeling accuracy while maintaining reasonable acquisition costs.
4Productivity
If existing building segmentation algorithms are used in dense urban areas, then processing speed is maintained, but segmentation accuracy deteriorates
Solution Approach 1:
The patent applies different processing strategies to different regions: in dense urban areas, it uses specialized algorithms that account for high building density and complex geometries, while in less dense areas, it can use faster but simpler methods. This local adaptation ensures high segmentation accuracy in dense regions without significantly compromising overall processing speed.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables efficient and accurate automatic extraction of 3D models for dense urban regions by improving building segmentation and clutter removal, ensuring consistent orientations and sharp boundaries between regions.
Implementation Method 1
Light Detection and Ranging (LIDAR) has emerged in recent years as a viable and cost-effective alternative to using solely 2D imagery, because LIDAR can directly produce precise and accurate range information
Data Source
AI summary
A method for extracting a 3D terrain model for identifying at least buildings and terrain from LIDAR data is disclosed, comprising the steps of generating a point cloud representing terrain and buildings mapped by LIDAR; classifying points in the point cloud, the point cloud having ground and non-ground points, the non-ground points representing buildings and clutter; segmenting the non-ground points into buildings and clutter; and calculating a fit between at least one building segment and at least one rectilinear structure, wherein the fit yields the rectilinear structure with the fewest number of vertices. The step of calculating further comprises the steps of (a) calculating a fit of a rectilinear structure to the at least one building segment, wherein each of the vertices has an angle that is a multiple of 90 degrees; (b) counting the number of vertices; (c) rotating the at least one building segment about an axis by a predetermined increment; and (d) repeating steps (a)-(c) until a rectilinear structure with the least number of vertices is found.