Vehicle Camera Image Correction for Distortion-Free Bird's-Eye Views
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle camera systems with manufacturing errors cannot generate accurate bird's-eye views due to distortion, leading to incorrect image representation, which affects driver safety and autonomous driving systems.
Innovation Solution
An image correction apparatus that acquires and corrects image data based on difference tables accounting for manufacturing errors, eliminating distortion and combining images from multiple cameras to produce an ideal bird's-eye view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If cameras with wide angle lenses are used to obtain bird's-eye view images, then the field of view is improved, but image distortion increases
Solution Approach 1:
The patent applies parameter changes by using different function models for different regions of the image. Specifically, it divides the image into multiple regions and applies different distortion correction functions to each region, allowing the system to maintain the wide field of view while correcting distortion through region-specific parameter adjustments
Solution Approach 2:
The patent segments the image into multiple regions and applies different function models to each region. This segmentation approach allows distinct distortion correction strategies to be applied to different parts of the image, resolving the contradiction between maintaining wide angle coverage and correcting distortion
2Measurement precision
If distortion correction is applied to images from cameras with manufacturing errors, then image accuracy is improved, but correction precision deteriorates due to unaccounted manufacturing variations
Solution Approach 1:
The patent applies local quality by assigning different function models to different regions of the image based on local characteristics. This allows the correction system to adapt to local manufacturing variations and optical distortions specific to each region, improving both image accuracy and correction precision simultaneously
3Area of stationary object
If multiple cameras are used to generate bird's-eye view, then coverage area is improved, but image combination accuracy deteriorates due to distortion inconsistencies
Solution Approach 1:
The patent uses parameter changes by applying region-specific function models to images from multiple cameras before combination. This ensures that distortion correction parameters are optimized for each camera's specific characteristics and viewing angle, enabling accurate image combination across the entire coverage area
Data Source
AI summary
An image correction device including: an image data acquisition part for acquiring the image data obtained by a camera; and a correction part for correcting the image data on the basis of a first difference that is a difference between a first image height obtained by light having entered the camera at a first incident angle and a first reference image height obtained by light having entered the camera at the first incident angle when the camera is free of manufacturing errors.