Traffic Signal Recognition Using Lamp Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic signal recognition systems face limitations in accuracy due to reliance on camera image analysis, which is influenced by camera performance and requires extensive learning to identify various shapes and patterns, leading to suboptimal recognition of traffic signal lighting states.
Innovation Solution
A traffic signal recognition system that combines camera image information with independent lamp pattern information, recognizing the lighting state by comparing detected traffic signal patterns with pre-defined lamp pattern data to improve accuracy and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera image analysis is used to recognize traffic signal lighting state, then the system can detect traffic signals, but the recognition accuracy is limited due to camera performance constraints and the need for extensive learning
Solution Approach 1:
The patent introduces an intermediary processing step that extracts lamp pattern information from detected traffic signals and compares it with pre-stored lamp pattern data. This intermediary approach bridges the gap between raw camera images and accurate lighting state recognition, reducing dependency on complex learning while improving precision through pattern matching
Solution Approach 2:
The system performs preliminary actions by pre-storing lamp pattern information corresponding to different lighting states before actual traffic signal detection. This pre-prepared reference data enables faster and more accurate recognition without requiring extensive real-time learning, thereby improving measurement precision while reducing device complexity
2Measurement precision
If only camera image information is used for recognition, then the system structure is simpler, but the recognition accuracy is insufficient due to influence from camera performance and environmental factors
Solution Approach 1:
The patent merges camera image information with lamp pattern information in a combined recognition process. By integrating multiple information sources (raw image data and extracted lamp patterns) and comparing them with pre-stored reference data, the system achieves higher recognition accuracy while maintaining a manageable system structure through systematic information fusion
3Measurement precision
If the system focuses on single detected part analysis, then the processing is faster, but the recognition accuracy is reduced due to unknown shapes and patterns of individual parts
Solution Approach 1:
The patent applies partial action by selectively analyzing specific lamp parts and their patterns rather than processing all detected elements in full detail. By focusing on key discriminative features and using pre-stored pattern data for comparison, the system achieves accurate recognition without the computational overhead of exhaustive single-part analysis, balancing precision and processing speed
Data Source
AI summary
Camera image information includes an image that is imaged by a camera installed on a vehicle. Lamp pattern information, which is information on a traffic signal having plural lamp parts, indicates a relative positional relationship between the plural lamp parts and an appearance of each lamp part when lighted. A system detects a subject traffic signal around the the vehicle based on the camera image information to acquire traffic signal detection information that indicates at least an appearance of each of plural detected parts of the subject traffic signal. The system compares the traffic signal detection information with the lamp pattern information. The system recognizes a lighting state of the plural lamp parts that is consistent with the appearance of each of the plural detected parts of the subject traffic signal as a lighting state of the subject traffic signal.