Ground Surface Detection Using Height-Distributed Point Clouds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for ground surface detection in three-dimensional scenes suffer from low accuracy, particularly in complex environments where the plane with the largest number of interior points may not be the actual ground surface.
Innovation Solution
A method and apparatus that involve acquiring an image acquisition position, determining a target point cloud within a preset range below this position, calculating the height distribution of this point cloud, and using this distribution to accurately identify the ground surface by analyzing the height differences and point densities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the terminal device determines the plane with the largest number of interior points as the ground surface, then the ground surface detection can be performed in simple environments, but the detection accuracy deteriorates in complex environments
Solution Approach 1:
The patent changes the detection parameter from simply counting interior points to analyzing height distribution characteristics. By examining the vertical distribution of point cloud heights and identifying peak density regions, the method adapts to complex environments while maintaining operational feasibility.
Solution Approach 2:
The patent transitions from two-dimensional plane fitting to three-dimensional height distribution analysis. By considering the vertical dimension (height) of point cloud data and analyzing density distribution across multiple height levels, the method achieves superior accuracy in complex scenes compared to traditional 2D plane detection.
2Device complexity
If the terminal device uses traditional plane fitting methods to detect ground surface, then the detection process is computationally simple, but the detection accuracy deteriorates when the environment is complex
Solution Approach 1:
The patent changes the computational approach from minimizing 2D plane fitting errors to analyzing 3D height distribution statistics. By calculating height frequencies and identifying peak density regions in the vertical dimension, the method achieves better accuracy without excessive computational burden.
Solution Approach 2:
The patent replaces the mechanical plane fitting process with a statistical height distribution analysis. Instead of iteratively adjusting plane parameters to fit points, the method uses height frequency counting and peak identification, which is computationally more efficient and more accurate for ground surface detection.
Data Source
AI summary
The present disclosure provides a ground surface detection method and apparatus, and a terminal device. The method includes: acquiring an image acquisition position in a target scene; determining a target point cloud from a point cloud related to the target scene according to the image acquisition position, wherein the target point cloud is a point cloud in a preset range below the image acquisition position in the target scene; determining a height distribution of the target point cloud in the target scene; and determining a position of a ground surface in the target scene according to the image acquisition position and the height distribution of the target point cloud.