Lane Division Line Recognition Using Dual Imaging Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional lane division line recognition systems face challenges in accurately recognizing the shape of road division lines, especially near the vehicle due to low resolution and environmental factors like dirty windshields and inclement weather, leading to recognition errors in offset distance and inclination.
Innovation Solution
A lane division line recognition apparatus using dual imaging units - a frontward unit for capturing images ahead and a periphery unit for capturing images of the vehicle's surroundings, with detection and reliability calculation units to determine the division line shape by combining or selecting between the two based on reliability scores, enhancing accuracy and robustness against environmental influences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a frontward imaging unit is used to capture images at a predetermined distance, then the imaging range is sufficient, but the resolution near the moving body becomes low
Solution Approach 1:
The patent divides the imaging function into two separate imaging units: a frontward imaging unit for capturing distant road surfaces and a periphery imaging unit for capturing near regions. This segmentation allows each unit to be optimized for its specific imaging range, with the periphery unit providing high-resolution images of the area near the moving body while the frontward unit maintains adequate imaging range for distant features.
2Device complexity
If only frontward imaging is used, then the system structure is simple, but recognition errors occur in offset distance and inclination
Solution Approach 1:
The patent merges the data from two imaging units (frontward and periphery) through a synthesis process. The periphery division line shape detection unit detects division lines in the near region with high precision, while the frontward unit covers distant areas. By combining these complementary data sources, the system achieves accurate recognition of offset distance and inclination without excessive structural complexity.
3Device complexity
If a single imaging unit is used, then the device is simple, but recognition accuracy drops in dirty windshield or inclement weather conditions
Solution Approach 1:
The patent changes the imaging parameters by using two imaging units with different positions and fields of view. When environmental conditions (dirty windshield, inclement weather) degrade the quality of images from one unit, the system can rely on or switch to data from the other unit, which may have different exposure conditions. This parameter diversity improves reliability without requiring a single overly complex imaging system.
Data Source
AI summary
The apparatus includes: a frontward imaging unit configured to capture an image ahead of the moving body; a frontward division line detection unit configured to detect a shape of a division line of a road surface, from the captured image, and calculate reliability of a frontward division line shape; a periphery imaging unit configured to capture an image of a periphery of the moving body; a periphery division line detection unit configured to detect a shape of the division line on the road surface, from the captured image, and calculate reliability of a periphery division line shape; and a division line shape determination unit configured to determine the shape of the division line on the road surface where the moving body moves, by selecting one from or combining the frontward division line shape and the periphery division line shape, based on first reliability and second reliability.