Parking Area Mapping from Vehicle Camera Streams and Space Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional parking spot availability solutions are costly to install and maintain, and do not provide a personalized experience for users, as they require modifications to the parking structure and rely on simple sensors/cameras.
Innovation Solution
A vehicle camera system processes video streams to extract feature vectors, determining vehicle description information and space information, which are used to generate a mapping of the parking area, including vehicle locations and available spaces, potentially using wireless connections for candidate information verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If simple sensors/cameras are used to detect parking spot availability, then the detection function is achieved, but the installation and maintenance cost increases due to modification of parking structure
Solution Approach 1:
The system uses the vehicle's own camera and processing capabilities to detect parking availability, rather than requiring the parking structure to provide detection infrastructure. The vehicle independently captures images, processes them to detect occupied spaces, and generates navigation information without external assistance.
Solution Approach 2:
Instead of having the parking structure actively detect and communicate occupancy status to vehicles, the system inverts the approach by having vehicles passively capture and process images to determine occupancy. The detection function moves from the static infrastructure to the moving vehicle.
2Adaptability or versatility
If traditional sensor/camera solutions are deployed, then parking spot detection is enabled, but the system cannot provide personalized experience for users
Solution Approach 1:
The system pre-processes images to extract features such as vehicle presence detection, space occupancy status, and generates navigation information before the user needs it. This preliminary processing enables personalized routing and parking recommendations when the user queries the system.
Solution Approach 2:
The system provides feedback to users in the form of personalized navigation information and parking recommendations based on detected occupancy status. The feedback loop allows the system to adapt to user needs and provide customized parking solutions rather than generic detection data.
3Loss of information
If video stream processing is used to extract vehicle description and space information, then personalized mapping is achieved, but processing time and computational resources increase
Solution Approach 1:
The system extracts only the essential features from video streams needed for parking detection, such as vehicle presence, occupancy status, and basic vehicle descriptions. By focusing on extracting only necessary information rather than processing all video data, the system reduces processing time while maintaining accuracy for navigation purposes.
Data Source
AI summary
In certain embodiments, a vehicle may obtain a video stream via a camera system of the vehicle. The video stream may be processed to extract one or more feature vectors from the video stream. Based on the feature vectors extracted from the video stream, vehicle description information related to one or more parked vehicles in the parking area may be determined. Space information indicating locations of the parked vehicles relative to available parking spaces in the parking area may also be determined based on the feature vectors extracted from the video stream. A mapping of the parking area may then be generated based on the vehicle description information and the space information.