Surround View Calibration Adjustment for Height Variations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surround view vision systems for vehicles face distortion issues when stitching images from cameras in environments with elevated surfaces, as the predefined geometric mapping data fails to provide a seamless and accurate top-down view, leading to distorted representations of the surrounding area.
Innovation Solution
The system adjusts calibration data by using sets of predefined calibration data records for different height levels in overlapping regions, allowing for tile-wise adjustments to minimize distortions and omissions, and incorporates height level information from sensors like ultrasonic or LIDAR to update the calibration data dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If predefined geometric mapping data from ground-level calibration is used for stitching, then the stitching process is simple and fast, but the representation becomes distorted in environments with elevated surfaces
Solution Approach 1:
The system dynamically adjusts calibration parameters based on detected height variations in the environment. Instead of using fixed ground-level calibration data, the stitching process adapts to different surface elevations by modifying mapping parameters in real-time, resolving the contradiction between simple stitching and accurate representation in varied terrains
Solution Approach 2:
The invention changes the calibration parameters from fixed ground-level values to height-adjusted values. By detecting the height of surfaces and correspondingly adjusting the calibration data, the system maintains both stitching efficiency and view accuracy across different environmental conditions
2Manufacturing precision
If height level adjustments are incorporated into calibration data, then view accuracy improves in elevated terrains, but the system complexity increases
Solution Approach 1:
The system performs preliminary height detection and calibration adjustment before the actual stitching process. By pre-adjusting calibration parameters based on detected height levels, the system achieves accurate stitching in elevated terrains without adding significant complexity to the main stitching workflow
Solution Approach 2:
The invention introduces height detection data as an intermediary element between the camera inputs and the stitching process. This intermediary information enables accurate calibration adjustment without fundamentally changing the core stitching mechanism, thereby managing system complexity while improving view accuracy
Data Source
AI summary
The present application relates to a system for generating a surround view and a method of operating the system. A synthesizer module synthesizes an output frame from input frames in accordance with predefined calibration data. The input frames have an overlapping region imaging an overlapping field of view captured by two adjacent cameras. An adjustment module receives height level information representative of a height level in the overlapping region; selects a data record out of a set of predefined calibration data records in accordance with the height level information; and updates the predefined a part of the calibration data with the selected data record.