Road Surface Boundary Detection via Spatial Projection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional technologies for detecting the boundary between an object and a road surface rely on the constant relationship between parallaxes and depths in multiple images, failing to accurately detect boundaries when this relationship is not maintained.
Innovation Solution
A detecting device comprising a first estimating unit for spatial position relationships, a second estimating unit for spatial position estimation, a third estimating unit for road surface position relationships, a projecting unit for calculating projection positions, and a calculator for updating object existence probabilities, allowing detection of boundaries even when parallaxes and depths are not in a constant relationship.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional technology uses a plurality of images captured by a plurality of cameras with the premise that parallaxes and depths are in a constant relationship, then the boundary detection can be performed under ideal conditions, but the detection accuracy deteriorates when this premise is not satisfied
Solution Approach 1:
The invention changes the fundamental parameter assumption from constant parallax-depth relationship to variable relationship. By estimating spatial positions and relationships without assuming constant parallax-depth correspondence, the system adapts to varying imaging conditions while maintaining detection accuracy
Solution Approach 2:
The invention makes the detection system dynamic by allowing the parallax-depth relationship to vary. Instead of using a fixed constant relationship, the system dynamically estimates spatial positions based on actual image data, enabling adaptation to different imaging scenarios
2Reliability
If multiple image capturing units are used to estimate spatial positions and detect boundaries, then detection capability is improved, but device complexity increases
Solution Approach 1:
The invention makes a single image capturing unit perform multiple functions: capturing images for spatial position estimation, calculating parallaxes, and detecting boundaries. This multi-functional approach maintains detection reliability while reducing the number of required capturing units
Solution Approach 2:
The system uses the captured images themselves to estimate spatial positions and relationships without requiring additional sensors or devices. The image data serves multiple purposes in the detection process, reducing overall system complexity
Data Source
AI summary
According to an embodiment, a detecting device includes a projecting unit, a calculator, and a detector. The projecting unit is configured to obtain a first projection position by projecting a capturing position of a captured image on a road surface, obtain a second projection position by projecting a spatial position in the captured image on the road surface, and obtain a third projection position by projecting an error position on the road surface. The calculator is configured to calculate an existence probability of an object on a line passing through the first and second projection positions so that an existence probability of the object between the second and third projection positions on the straight line is greater than between the first and third projection positions on the straight line. The detector is configured to detect a boundary between the road surface and the object by using the existence probability.