Stop Line Recognition Device Reducing Calculation Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing stop line recognition systems generate excessive calculation load and reduce accuracy due to unnecessary recognition and misrecognition of shadows as stop lines, especially when traffic signals are not directly relevant for vehicle stop control.
Innovation Solution
A stop line recognition device that captures images of the vehicle's environment, recognizes traffic signals, and only initiates stop line recognition when a red traffic signal is detected, using a monocular camera and image processing to set a stop line recognition frame and determine luminance changes within specific regions, thereby reducing unnecessary calculations and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If stop line recognition is constantly performed at a level similar to preceding vehicle and traffic signal recognition, then the stop line can be recognized continuously, but unnecessary calculation load increases and processing time becomes excessively long
Solution Approach 1:
The system performs stop line recognition periodically based on traffic signal states rather than continuously. Stop line recognition is executed only when a red traffic signal is detected, and suspended when green light is detected, creating a periodic recognition pattern that reduces calculation load while maintaining necessary recognition accuracy
Solution Approach 2:
The stop line recognition function is made dynamic by enabling it only under specific conditions (red traffic signal detection) and disabling it under other conditions (green light detection). This dynamic activation based on traffic signal state allows the system to adapt recognition intensity to actual needs, balancing accuracy and processing efficiency
2Reliability
If stop line recognition is constantly performed, then the stop line can be recognized without delay, but recognition accuracy decreases due to misrecognition of shadows and light-dark patterns as stop lines
Solution Approach 1:
The system uses traffic signal detection results as feedback to control stop line recognition execution. When red light is detected, stop line recognition is activated; when green light is detected, it is suspended. This feedback mechanism ensures recognition occurs only when relevant, improving both reliability and accuracy by avoiding false recognition of shadows during green light periods
Solution Approach 2:
The system performs preliminary traffic signal detection before executing stop line recognition. By detecting the traffic signal state in advance and using it as a prerequisite condition, the system ensures that stop line recognition is only performed when necessary, preventing premature or unnecessary recognition that could lead to shadow misidentification
Data Source
AI summary
An image processing unit 4 performs the recognition of a traffic signal in front of an own vehicle on the carriageway thereof. When the image processing unit 4 recognizes that there is a traffic signal front closet to the vehicle on the carriageway within a set distance L2 and the traffic signal is displaying red light, the process shifts to a stop line recognition mode and the image processing unit 4 performs stop line recognition. Accordingly, unnecessary execution of stop line recognition can be accurately avoided. Therefore, a stop line that is necessary for the control of the own vehicle can be recognized with a high accuracy without generating excessive calculation load.