Vehicle Camera Distortion Correction via Feature Topology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wide-angle cameras, especially those with fish-eye lenses, produce severely distorted images due to their short focal length and wide field of view, making it difficult to achieve perfect rectilinearity without accurate distortion correction factors and formulas, which are often compromised by mechanical errors in sensor-lens coupling.
Innovation Solution
An image processing device and method that detect representative feature points and topology information from input images, correct distortion using homography, and determine an optimal light center for each camera, allowing for accurate calibration and image matching to generate a top-view image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a wide-angle lens with short focal length is used to achieve wide field of view, then the field of view is widened, but severe radial distortion occurs in the captured image
Solution Approach 1:
The patent applies preliminary action by pre-calculating distortion correction values for each pixel position based on the camera's intrinsic parameters (focal length, principal point, distortion coefficients). A correction lookup table is generated in advance, storing the mapping between distorted pixel coordinates and corrected coordinates. During image capture, this pre-computed correction data is applied to rectify radial distortion without requiring real-time complex calculations, thus maintaining both wide field of view and image rectilinearity.
2Manufacturing precision
If standard distortion correction formulas are used with manufacturer-provided parameters, then correction can be applied, but mechanical errors in sensor-lens coupling cause inaccurate light center values leading to residual distortion
Solution Approach 1:
The patent implements feedback by using captured test images to actually measure the principal point and distortion coefficients of each specific camera unit, rather than relying on theoretical manufacturer specifications. The system captures images of a calibration pattern, detects feature points, and iteratively optimizes the distortion parameters to minimize residual distortion. This measured data feeds back into the correction process, ensuring high accuracy despite mechanical variations in sensor-lens coupling.
3Manufacturing precision
If manual calibration of each camera is performed to achieve accurate distortion correction, then rectilinearity is improved, but production time and costs increase
Solution Approach 1:
The patent applies self-service by enabling each camera module to automatically calibrate itself during the assembly process. The calibration system captures images autonomously, processes them through the distortion correction algorithm, and generates correction parameters specific to each camera without requiring manual intervention. This automated self-calibration approach maintains high rectilinearity accuracy while dramatically reducing production time and labor costs compared to manual calibration methods.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Disclosed are an image processing device and an image processing method. The image processing device includes: an image obtaining unit for obtaining input images captured by a plurality of cameras mounted in a vehicle; and a controller for detecting a representative feature point representing a shape feature of a particular pattern included in the input images, detecting topology information with respect to corner points of the particular pattern based on the detected representative feature point, and determining an optimal light center corresponding to each of the plurality of cameras based on the detected topology information.