Bird's-Eye-View Obstacle Detection via Image Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional bird's-eye-view imaging systems struggle with detecting obstacles of varying heights due to image stitching issues and fixed visual angles, which can lead to incomplete obstacle detection and increased system costs with the use of additional sensors like radars or ultrasonic devices.
Innovation Solution
The system generates bird's-eye-view images with overlapped regions, using a correlation coefficient analysis to determine obstacle presence and position without additional sensors, and adjusts visual angles based on steering-wheel-turning-angle signals to provide flexible and accurate obstacle detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional planar images are used for vehicle driving assistance, then the system cost is low, but the visual dead angle is large and obstacle detection is incomplete
Solution Approach 1:
The patent transforms conventional 2D planar images into a 3D bird's-eye-view perspective by processing images from multiple cameras positioned around the vehicle. This dimensional transformation allows the system to overcome the limitations of planar images by providing a top-down view that eliminates visual dead angles and enables complete obstacle detection across all surrounding areas.
2Measurement precision
If additional sensors like radars or ultrasonic devices are added to detect obstacles in dead angle regions, then obstacle detection accuracy is improved, but system cost increases
Solution Approach 1:
The patent makes the camera system multi-functional by using it not only for visual display but also for obstacle detection through image correlation analysis. By processing bird's-eye-view images from multiple cameras and analyzing correlations between overlapping regions, the system achieves obstacle detection functionality that would otherwise require separate sensors, thereby avoiding additional system cost while maintaining detection accuracy.
Solution Approach 2:
The patent introduces image correlation coefficient as an intermediary mechanism to detect obstacles. Instead of directly using additional sensors, the system uses the correlation analysis between overlapping regions of bird's-eye-view images from adjacent cameras as a mediator to identify obstacles, achieving accurate detection without the need for expensive additional sensing hardware.
3Adaptability or versatility
If a fixed virtual viewpoint is used in bird's-eye-view imaging, then the system is simple to implement, but the field of view cannot adapt to different driving conditions
Solution Approach 1:
The patent implements a dynamic virtual viewpoint system where the bird's-eye-view imaging position and field of view can be adjusted based on driving conditions. The system dynamically changes the virtual viewpoint location and adjusts which camera images are selected and how they are processed, allowing the field of view to adapt to different driving scenarios such as forward movement, backward movement, and cornering, while maintaining system simplicity through software-based control.
Data Source
AI summary
A technology of determining obstacles around a vehicle through utilizing bird's-eye-view images; wherein, a plurality of image fetching devices disposed in various positions of said vehicle fetch a plurality of images around said vehicle, said images of two adjacent regions contain at least an overlapped region; an image processor transforms said images into said respective independent bird's-eye-view images; and an obstacle detection unit compares said overlapped region in said independent bird's-eye-view images of two adjacent regions, so as to obtain their correlations, and existence of said obstacle is determined based on said correlations. Moreover, a correspondence table is set up containing a set of space transformation information based on vehicle driving condition information. Therefore, a surrounding bird's-eye-view image of an appropriate visual angle can be produced quickly and a position of said obstacle is marked on said surrounding bird's-eye-view image.