Line Marking Identification Using Flare-Adaptive Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing line marking identification apparatuses struggle to accurately identify lane markings in images captured by vehicle-mounted cameras, especially at night when strong light from oncoming vehicles causes flare, leading to misrecognition of line markings due to exclusion of flare areas from recognition processing.
Innovation Solution
An apparatus that extracts paint candidates from images and determines whether they possess predefined flare features, setting a more stringent identification condition for flare candidates to include them in line marking identification processing, allowing for accurate identification even in flare-prone conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If flare areas are excluded from recognition processing to prevent misidentification, then reliability of line marking identification is improved, but productivity of line marking identification deteriorates due to loss of potentially valid line marking information
Solution Approach 1:
The invention changes the identification parameters by setting different identification conditions for flare paint candidates versus non-flare paint candidates. For flare paint candidates, a more stringent identification condition is applied, which adjusts the threshold parameters for determining whether a paint candidate represents an actual line marking. This allows the system to maintain high reliability for flare-affected areas while still detecting valid line markings that meet the stricter criteria.
2Reliability
If many flare areas are present in captured image, then reliability of preventing misidentification is improved, but measurement precision of line marking detection deteriorates due to reduced recognition processing areas
Solution Approach 1:
The invention applies local quality by implementing different identification conditions in different regions of the image based on flare detection. Areas identified as flare paint candidates receive a more stringent identification condition, while non-flare areas use standard identification conditions. This localized approach allows the system to maintain high precision in non-flare regions while applying appropriate filtering in flare-affected regions.
Data Source
AI summary
An apparatus for identifying a line marking on a road surface. In the apparatus, an extractor extracts a paint candidate that is a candidate for road surface paint used to identify a line marking in an image captured by a camera mounted on a vehicle to capture an image of an area including a road surface ahead of the vehicle. A determiner determines whether or not the paint candidate has at least one predefined flare feature. A line marking identifier identifies a line marking using the paint candidate which meets an identification condition used to identify a line marking. The line marking identifier sets the identification condition to be more stringent for a flare paint candidate that is the paint candidate determined by the determiner to have the at least one predefined flare feature than for the paint candidate determined by the determiner to have no predefined flare feature.