Surround View Parking Detection via Camera Motion Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic parking assist systems rely on ultrasonic detection, which struggles to utilize visual cues like lanes and signs and has a limited range, making them ineffective in situations with cars parked on either side, hindering the detection of vacant parking spots.
Innovation Solution
The system uses a combination of camera images from multiple cameras to simulate a surround view, estimates motion parameters, and tracks potential parking spaces using camera motion estimation, determining location, free space, and entrance corners, while clustering and selecting suitable parking spots based on distance thresholding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ultrasonic detection is used for parking spot detection, then the system can detect obstacles, but the detection range is limited and visual cues like lanes and signs cannot be utilized
Solution Approach 1:
The patent replaces ultrasonic detection (acoustic field) with optical detection using cameras (optical field). This substitution enables the system to achieve both long detection range and the ability to recognize visual cues such as lanes, signs, and parking spot markings, thereby resolving the limitation of ultrasonic detection range while maintaining obstacle detection capability.
2Reliability
If ultrasonic detection is used, then obstacle detection is possible, but the system cannot effectively detect vacant parking spots when cars are parked on either side
Solution Approach 1:
By substituting ultrasonic detection with optical detection using multiple cameras, the system gains the ability to recognize visual features such as lane markings, parking spot lines, and surrounding vehicles. This enables reliable detection of vacant parking spots even in constrained environments where cars are parked on both sides, as the optical system can interpret spatial relationships and identify available spaces through image processing.
Solution Approach 2:
The patent transitions from one-dimensional ultrasonic distance measurement to two-dimensional optical imaging. By capturing images from multiple cameras and stitching them into a composite view, the system creates a comprehensive spatial representation that enables detection of parking spots in complex, limited spaces where ultrasonic detection would fail.
3Length of stationary object
If radar detection is used, then detection range is improved, but the system discriminates distance rather than exploiting visual cues
Solution Approach 1:
The patent substitutes radar detection (electromagnetic field for distance measurement) with optical detection using cameras. This substitution maintains long detection range while adding the critical capability to recognize visual cues such as lanes, signs, and parking spot markings through image processing and pattern recognition algorithms.
Data Source
AI summary
A method of parking lot tracking including receiving a plurality of camera images from a plurality of cameras attached to a vehicle in motion, stitching the plurality of camera images to simulate a surround view of the vehicle, recognizing at least one potential parking space within the surround view of the vehicle, estimating a motion parameter by camera motion estimation of the plurality of cameras and tracking the at least one potential parking space based on the motion parameter.