Vehicle Parking Detection Using Front and Rear Camera Confidence Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing parking detection systems in vehicles require surround-view camera systems with multiple cameras, which are complex and expensive, making them unsuitable for entry-level and many middle-class vehicles.
Innovation Solution
Utilizing a front and rear camera system, including a monochrome front camera for advanced driver assistance and a color rear backup camera, to detect and confirm parking locations through image processing and confidence scoring, without the need for a surround-view system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a surround-view camera system with at least four wide field of view cameras is used for parking detection, then parking detection accuracy is improved, but device complexity and cost increase
Solution Approach 1:
The patent reuses existing front and rear cameras that serve other primary functions (ADAS and backup viewing) for the additional purpose of parking detection. This multi-functionality approach eliminates the need for dedicated surround-view cameras while maintaining parking detection capability, thereby reducing device complexity and cost without sacrificing detection accuracy
Solution Approach 2:
The system leverages camera infrastructure that is already present in the vehicle for other safety and assistance functions. By making these existing cameras serve dual purposes including parking detection, the system avoids adding separate dedicated hardware, thus reducing overall system complexity while maintaining functionality
2Measurement precision
If a surround-view camera system with at least four wide field of view cameras is used for parking detection, then parking detection accuracy is improved, but system cost increases
Solution Approach 1:
The patent reuses existing front and rear cameras that serve other primary functions (ADAS and backup viewing) for the additional purpose of parking detection. This multi-functionality approach eliminates the need for dedicated surround-view cameras while maintaining parking detection capability, thereby reducing device complexity and cost without sacrificing detection accuracy
Solution Approach 2:
The system uses standard, mass-produced front and rear cameras that are already included in most modern vehicles for other purposes. These conventional cameras are more cost-effective than specialized wide field of view surround-view cameras, reducing the overall system cost while achieving the required parking detection functionality
3Area of stationary object
If multiple wide field of view cameras are installed for parking detection, then coverage area is improved, but device complexity increases
Solution Approach 1:
The patent utilizes the temporal dimension by capturing multiple images at different positions along the roadway. Instead of relying solely on multiple simultaneous cameras, the system achieves comprehensive coverage by sequentially capturing images as the vehicle moves, processing them to identify parking locations. This dimensional approach reduces hardware complexity while maintaining coverage area
Data Source
AI summary
A parking detection method includes: accessing data corresponding to an image from a camera on a vehicle; computing a confidence score for a possible vacant parking location from the data corresponding to the image from the camera on the vehicle; and when the confidence score is greater than a threshold confidence, identifying the possible vacant parking location as an actual vacant parking location from the data corresponding to the image from the camera on the vehicle.