Front Curb View Distortion Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Drivers face challenges in accurately judging the position of a vehicle relative to obstacles like curbs during parking due to the obstruction from the vehicle's front, leading to potential damage from hitting the curb, as existing forward-view camera systems do not provide optimized, distortion-corrected images for precise positioning.
Innovation Solution
A system and method for creating an enhanced perspective view using images from left-front and right-front cameras, employing a virtual image surface model and processing techniques to correct distortions such as stretching and double imaging, providing a clear and accurate representation of the curb or other frontal objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of moving object
If wide-angle lens images are used to capture the area in front of the vehicle, then the field of view is increased, but distortion and exaggerated perspective effects are introduced
Solution Approach 1:
The patent applies parameter changes by transforming the imaging model from a traditional pinhole camera model to a catadioptric camera model that accounts for the wide-angle lens distortion characteristics. This allows the system to capture a broader field of view while compensating for distortion through mathematical transformation of the imaging parameters, thereby maintaining measurement precision despite using wide-angle optics.
2Measurement precision
If rectilinear projection is used to de-warp wide-angle images, then perspective distortion is corrected, but stretching effects appear in peripheral areas
Solution Approach 1:
The patent applies local quality by implementing different projection strategies for different regions of the image. The system uses rectilinear projection for central areas where accurate perspective correction is needed, while applying alternative warping techniques to peripheral regions to minimize stretching effects. This region-specific approach allows the system to optimize for both distortion correction and shape preservation in different parts of the image.
3Area of moving object
If left-front and right-front camera images are merged to create a composite view, then coverage area is increased, but double imaging of objects occurs in overlap regions
Solution Approach 1:
The patent applies feedback by implementing a seamless blending mechanism that uses overlap region analysis to detect and eliminate double imaging. The system continuously monitors the transition zones between left-front and right-front camera images, and dynamically adjusts the blending weights based on detected features and distortion patterns. This feedback-driven approach ensures that objects in overlap regions appear only once in the composite view while maintaining full coverage area.
Data Source
AI summary
A system and method for creating an enhanced perspective view of an area in front of a vehicle, using images from left-front and right-front cameras. The enhanced perspective view removes the distortion and exaggerated perspective effects which are inherent in wide-angle lens images. The enhanced perspective view uses a camera model including a virtual image surface and other processing techniques which provide corrections for two types of problems which are typically present in de-warped perspective images—including a stretching effect at the peripheral area of a wide-angle image de-warped by rectilinear projection, and double image of objects in an area where left-front and right-front camera images overlap.