Road Surface Point Cloud Completion for Recess Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately detect recessed portions on a road surface using camera-mounted mobile bodies, as it is difficult to differentiate between a recessed portion and a non-recessed portion based on image data.
Innovation Solution
An electronic device that acquires point cloud data, discriminates missing parts on the road surface, complements these parts with pseudo point clouds, and determines recessed portions using a cloth simulation algorithm to enhance detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a camera-mounted mobile body is used to detect road surface features, then the system can capture image data of the road surface, but it becomes difficult to accurately determine recessed portions from the image data
Solution Approach 1:
The patent replaces the optical imaging system (camera) with a laser-based point cloud detection system. The laser detector emits laser beams to the road surface and measures reflected light to generate precise three-dimensional point cloud data, substituting the mechanical/optical image capture method with a laser ranging method that directly measures distance and elevation information.
Solution Approach 2:
The patent transitions from two-dimensional image data to three-dimensional point cloud data by adding depth information through laser ranging. This dimensional enhancement allows the system to detect recessed portions by comparing elevation values across multiple points in three-dimensional space, making it possible to identify depth variations that are indistinguishable in flat images.
2Loss of information
If point cloud data is acquired using a laser detector, then three-dimensional road surface information can be obtained, but missing parts in the point cloud data cannot be detected
Solution Approach 1:
The patent performs preliminary detection of missing parts in the point cloud data before final analysis. The processor identifies regions where point cloud data is absent or insufficient, flags these as missing parts, and applies targeted processing to generate virtual point clouds for these specific regions, ensuring no information is lost in the detection process.
Solution Approach 2:
The patent introduces virtual point clouds as an intermediary element to bridge the gap between detected and missing road surface information. These virtual point clouds are generated through cloth simulation and serve as mediators that fill in missing data, allowing the system to reconstruct complete road surface topology even when direct measurement data is absent.
3Measurement precision
If missing parts are not complemented, then processing is simpler, but recessed portions cannot be accurately determined
Solution Approach 1:
The patent applies local quality processing by treating missing parts differently from complete regions. Instead of uniform processing across the entire point cloud, the system identifies specific missing regions and applies cloth simulation only to those localized areas, generating virtual point clouds with appropriate depth characteristics tailored to each missing region's context.
Solution Approach 2:
The patent segments the point cloud processing into distinct stages: detection of missing parts, generation of virtual point clouds for missing regions, and final integration. This segmentation allows the complex task of handling incomplete data to be broken down into manageable steps, reducing overall processing complexity while maintaining accuracy.
Data Source
AI summary
An electronic device (100) includes an acquiring unit (131) configured to acquire point cloud data (123) indicating a set of points capable of identifying a state of a road surface, a discriminator (132) configured to discriminate a missing part of the point cloud data (123) which is located at a height equal to or lower than a height of the road surface in a detection range of the point cloud data (123), a complementor (133) configured to complement the missing part with a pseudo point cloud at a position of a predetermined depth from the road surface, and a determiner (135) configured to determine a recessed portion of the road surface based on the point cloud data (123) obtained by complementing the pseudo point cloud.