Surround View Calibration Using Asymmetric Markers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Surround view systems face calibration errors due to reflections of marking areas in vehicle components and external objects, leading to unreliable image stitching and display of vehicle surroundings.
Innovation Solution
A surround view system that uses a control device to distinguish between real and mirrored axis-asymmetrical or point-asymmetrical marking areas, only using unmirrored areas for calibration to ensure accurate image merging and reduce errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If marking surfaces are used for calibration, then calibration can be performed, but reflections of marking surfaces cause calibration errors
Solution Approach 1:
The marking surface incorporates an axially asymmetric and/or point-asymmetric area that serves as a mirror image recognition feature. This asymmetric pattern allows the control unit to distinguish between the real marking surface and its reflections by detecting whether the asymmetric pattern is mirrored. The asymmetric feature is integrated into the marking surface design, enabling automatic differentiation between direct and reflected images during calibration.
2Measurement precision
If the control unit searches for all marking surfaces, then more calibration data is available, but mirrored marking surfaces introduce errors
Solution Approach 1:
The control unit performs feedback analysis by detecting the asymmetric pattern in captured images and determining whether it represents a mirrored or non-mirrored marking surface. Based on this detection, the control unit selectively uses only the non-mirrored marking surfaces for calibration. This feedback mechanism ensures that reflected images are automatically excluded from calibration data, maintaining high measurement precision without losing valid calibration information.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enhances the reliability and accuracy of surround view system calibration by excluding reflections from vehicle components and external objects, resulting in a more precise and error-prone all-round vision system.
Implementation Method 1
The control unit is designed to search for the marker surface in both the first and second images for calibration purposes and to use only those marker surfaces found in the first and second images whose axially asymmetric and/or point-asymmetric areas are not mirrored compared to the marker surface of the vehicle's surroundings
Implementation Method 2
A camera calibration device, a vehicle equipped with a plurality of cameras, and a camera calibration method are disclosed in US 2010/0194886 A1. The camera calibration method verifies an arrangement state of a calibration pattern.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The invention relates to a surround-view system (101) for a vehicle (100), comprising a first camera (103) for capturing a first image (110) of the vehicle's surroundings (107), which has a marker area (108) with an axis-asymmetric and/or point-asymmetric region (109), and a second camera (104) for capturing a second image (111) of the vehicle's surroundings, which has the marker area. Furthermore, the surround-view system (101) includes a control unit (106) for combining the first and second images and for calibrating the surround-view system, wherein the control unit is configured to search for the marker area in both the first and second images for calibration purposes.The control unit is further designed to use only those marking areas found in the first and second images for calibrating the surround-view system whose axis-asymmetric and/or point-asymmetric areas are not mirrored compared to the marking area of the vehicle environment.