Aerial Line Extraction Using Point Cloud Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing aerial line extraction systems face difficulties in distinguishing aerial lines from trees in three-dimensional point cloud data, leading to noise interference and inaccurate model estimation, especially in mountainous areas.
Innovation Solution
An aerial line extraction system that includes an area-of-interest cropping unit, an aerial line candidate extraction unit, and an aerial line model estimation unit, which segments three-dimensional point cloud data by slice planes, clusters regions, and classifies clusters by size to separate aerial lines from noise like trees, using utility pole coordinates as references.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If noise removal is performed by removing unnatural catenary curves from three-dimensional point cloud data, then aerial lines can be extracted, but trees become noise in mountainous areas and a large number of catenary curves become candidates making it difficult to remove noise
Solution Approach 1:
The patent divides the three-dimensional point cloud data into multiple two-dimensional cross-sectional views at regular intervals along the extension direction. By processing each cross-sectional view separately and extracting candidate points that satisfy specific conditions (being candidate points in adjacent cross-sectional views and having appropriate distances from utility pole centers), the system can effectively distinguish aerial lines from tree noise without requiring complex noise removal algorithms.
Solution Approach 2:
The patent applies different extraction conditions to different spatial locations. Candidate points must satisfy location-specific conditions including being within a predetermined distance range from utility pole centers and appearing in adjacent cross-sectional views. This local quality approach allows the system to accurately identify aerial lines while filtering out tree noise in mountainous areas.
2Loss of information
If three-dimensional model data is generated from three-dimensional point cloud data and superimposed on image data to visualize facilities, then outdoor facilities can be detected, but it is difficult to separate aerial lines from trees in areas with many trees
Solution Approach 1:
The patent segments the three-dimensional point cloud data into multiple two-dimensional cross-sectional views. By extracting candidate points from each cross-sectional view and requiring them to appear in adjacent views, the system maintains facility detection capability while improving aerial line identification accuracy through the segmented processing approach.
Solution Approach 2:
The patent transforms the three-dimensional point cloud data into multiple two-dimensional cross-sectional views. This dimensionality change allows the system to extract aerial lines by analyzing patterns across multiple two-dimensional planes, making it easier to distinguish aerial lines from trees compared to processing the full three-dimensional data directly.
Data Source
AI summary
Provided is an aerial line extraction system including: an area-of-interest cropping unit that crops a region where a point cloud data of aerial lines is assumed to exist as an area of interest by setting coordinates of utility poles as a reference from a three-dimensional point cloud data of a three-dimensional shape that includes the aerial lines and trees installed in the air via the utility poles; an aerial line candidate extraction unit that extracts a candidate point cloud data of the aerial lines from the three-dimensional point cloud data in the area of interest; and an aerial line model estimation unit that estimates a model of the aerial lines on the basis of the extracted candidate point cloud data of the aerial lines.