Inverse Perspective Transform for Autonomous Parking Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current parking assistance systems struggle to recognize parking slots and obstacles, especially those adjacent to traffic lanes, curbs, or pedestrian areas, and fail to differentiate between types of parking slots, such as reserved spaces for the disabled, which is crucial for unmanned autonomous vehicles.
Innovation Solution
A device and method utilizing a wide-angle camera, in-vehicle sensors, and image processing techniques to perform inverse perspective transformation, detect obstacles and parking slot markings, build a line marking map, and recognize parking spaces, enabling autonomous parking by combining space and parking slot recognition technologies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of moving object
If a wide-angle camera is used to capture parking slot markings and obstacles, then the field of view is improved, but image distortion increases making accurate recognition difficult
Solution Approach 1:
The system dynamically adjusts the image processing approach based on the detected scene geometry. By performing inverse perspective transformation, the system adapts the distorted wide-angle image to a corrected coordinate system that preserves geometric relationships, enabling accurate recognition while maintaining the broad field of view advantage
Solution Approach 2:
The patent transforms the image coordinate system parameters through inverse perspective transformation. This mathematical transformation changes the geometric parameters of the image to correct the distortion inherent in wide-angle lenses, allowing accurate detection of parking slot markings and obstacles while retaining the expanded field of view
2Difficulty of detecting and measuring
If traditional space recognition methods using ultra sonic sensors are used, then the space detection capability is improved, but the ability to recognize parking slot types and markings is insufficient
Solution Approach 1:
The system merges space recognition capabilities with parking slot marking recognition by integrating multiple processing modules. The obstacle detection unit identifies spatial boundaries while the line marking detection unit simultaneously recognizes parking slot types, combining both functions into a unified system that prevents information loss
Solution Approach 2:
The vision-based system performs multiple functions simultaneously: it detects obstacles, recognizes parking slot markings, identifies parking slot types (including disabled parking), and determines available spaces. This multi-functional approach replaces the need for separate sensor systems while preserving all necessary information
3Measurement precision
If motion stereo vision technology is used for space recognition, then the depth perception is improved, but the recognition of parking slot markings and types remains insufficient
Solution Approach 1:
The system combines motion stereo vision for depth perception with inverse perspective transformation and line marking detection for parking slot recognition. By merging these processing streams, the system maintains accurate depth information while simultaneously extracting parking slot type and marking information that would be lost in pure stereo vision approaches
Data Source
AI summary
Provided herein is a device and a method for recognizing an obstacle and a parking slot to support an unmanned autonomous parking function. The device includes a motion measurement unit measuring a vehicle motion using an in-vehicle sensor, an inverse perspective transform unit performing inverse perspective transformation of an image, which is obtained using a wide-angle camera, to obtain an inverse perspective image, and an obstacle detection unit detecting the obstacle using the inverse perspective image.