Parking Space Recognition via Overlap and Brightness Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic parking systems relying on parking space line marking recognition are limited in urban environments where markings are often absent, requiring user input and suffering from high error rates during perpendicular parking, especially when using ultrasonic sensors.
Innovation Solution
A method for automatically recognizing parking spaces using a processor to detect and calculate overlap coefficients and brightness values from images, selecting the final parking space based on these calculations, and reflecting movement between images to accurately identify and differentiate parking spaces without user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If parking space line marking recognition is used, then target position accuracy is improved, but the system cannot operate when markings are absent
Solution Approach 1:
The system segments the parking space detection task into two independent components: line marking detection and free-space detection. This allows the system to process each type separately and combine results, enabling operation in both marked and unmarked environments without compromising accuracy in either scenario.
Solution Approach 2:
The system implements multi-functionality by integrating multiple detection algorithms (line marking recognition and free-space detection based on ultrasonic sensor data) into a single unified system. This universal approach allows the system to adapt to different parking environments (marked or unmarked) and select the appropriate detection method based on available infrastructure.
2Ease of manufacture
If ultrasonic sensor based free-space process is used, then the system is simple to implement, but error rate during perpendicular parking is unreasonably high
Solution Approach 1:
The system merges ultrasonic sensor-based free-space detection with line marking recognition algorithms. By combining these two methods, the system maintains the simplicity of ultrasonic sensor implementation while adding the accuracy of visual recognition for perpendicular parking scenarios, thereby reducing error rates without significantly complicating the system architecture.
3Ease of manufacture
If user interface based process is used, then the system is easy to implement, but user operation is required every time
Solution Approach 1:
The system implements self-service by automatically detecting parking spaces through line marking recognition and free-space analysis without requiring user input. The processor autonomously processes imaging device data, identifies suitable parking spaces, and provides target position information, eliminating the need for repeated user operations while maintaining ease of implementation through algorithmic automation.
4Measurement precision
If additional sensors like scanning laser radar or stereo camera are used, then detection accuracy is improved, but device complexity increases
Solution Approach 1:
The system uses image processing algorithms as an intermediary to extract parking space information from existing imaging device data. Instead of adding multiple sensors, the system employs computational methods (line segmentation, feature point detection, geometric relationship analysis) to process images and achieve accurate detection, thereby maintaining detection precision while avoiding increased device complexity.
Data Source
AI summary
Disclosed herein is a method for recognizing a parking space line marking for a vehicle, including: detecting, by a processor, a plurality of parking spaces from a portion of a parking space line marking in an image; calculating, by the processor, an overlap coefficient representing a degree of overlapping between the detected parking spaces; selecting, by the processor, a parking space having a largest brightness coefficient as a final parking space by determining overlap when the overlap coefficient has a predetermined magnitude and comparing the brightness degrees of the overlapped parking spaces.