Road Boundary Detection Using Scoring Hough Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle safety systems face challenges in precisely detecting road boundaries due to obstacles like other vehicles, which can lead to incorrect lane recognition and reduced reliability.
Innovation Solution
A road boundary detection system utilizing an optical scanner and the Scoring Hough Transform algorithm to generate a parameter space, prioritize straight lines based on contact points, and assign scores to determine the road boundary, effectively differentiating between contact and non-contact points to improve detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional road boundary detection methods are used, then the system is simple to implement, but the detection precision deteriorates when obstacles are present
Solution Approach 1:
The detection process is segmented into multiple stages: data acquisition from optical scanner, parameter space generation, straight line extraction, scoring, and road boundary determination. Each stage processes specific aspects of the data independently, allowing complex detection to be broken down into manageable steps that improve precision without overwhelming system complexity
Solution Approach 2:
The patent transforms the detection problem from direct image space analysis to parameter space analysis using Hough Transform. By converting contact points into parameter space coordinates (ρ, θ) and analyzing straight lines in this transformed dimension, the system achieves more robust boundary detection that is less sensitive to obstacles in the original image space
2Reliability
If obstacles are present in the detection field, then the detection coverage increases, but the reliability of road boundary recognition deteriorates
Solution Approach 1:
The patent introduces an intermediary scoring mechanism that evaluates the credibility of detected straight lines. The score calculator assigns scores based on the number of contact points supporting each line and penalizes lines that pass through non-contact points (obstacles). This intermediary scoring layer filters out unreliable detections caused by obstacles before final road boundary determination
Solution Approach 2:
The system implements feedback through the scoring mechanism that continuously evaluates detected straight lines against the contact point data. Lines that do not align well with actual contact points receive lower scores and are discarded, while high-scoring lines that consistently match contact patterns are selected as reliable road boundaries. This feedback loop ensures obstacles don't compromise final detection reliability
3Measurement precision
If more contact points are used to define straight lines, then the detection accuracy improves, but the computational complexity increases
Solution Approach 1:
The patent extracts multiple candidate straight lines from the parameter space, potentially more than strictly necessary. The scoring mechanism then evaluates these candidates and selects only the high-scoring ones. This approach of generating excessive candidates followed by selective filtering achieves high accuracy through thorough analysis while managing computational load through efficient scoring and selection
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
The system enhances the accuracy and reliability of road boundary detection even in the presence of obstacles, providing precise lane recognition and improving vehicle safety systems.
Implementation Method 1
an optical scanner configured to emit a light to an object to acquire measurement data reflected from the object
Data Source
AI summary
A road boundary detection system includes: an optical scanner configured to emit a light to an object to acquire measurement data reflected from the object; and a processor configured to extract a plurality of straight lines in order of a priority on each of the plurality of straight lines when a higher priority is given to a straight line including more of the contact points, based on the measurement data, configured to calculate a score of each straight line by assigning a score to the contact point included in the extracted straight lines and by assigning a score to a non-contact point included in other straight line having the same angle as the contact point, and configured to select a road boundary according to the priority of the calculated score of each straight lines.