On-Board Camera Roll Calibration Using Lane-Line Frame Sequences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for calibrating the roll angle of an on-board camera using inverse perspective mapping (IPM) on a single frame image result in inaccurate roll angle measurements.
Innovation Solution
A method that utilizes an image frame sequence collected by the on-board camera while the vehicle is in motion, selects image groups, obtains lane line images at a target roll angle, and determines the roll angle based on lane line image sets using big data statistics and fitting techniques to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If IPM projection is performed on a single frame image to fit lane line, then the processing speed is fast, but the roll angle calibration accuracy deteriorates
Solution Approach 1:
The patent segments the image processing into multiple groups of consecutive frames rather than processing a single frame. Each group contains multiple frames that are processed together to fit lane lines and calculate roll angles, dividing the complex task into manageable segments that improve accuracy through statistical aggregation
Solution Approach 2:
The patent merges multiple consecutive image frames into groups and combines the lane line fitting results from these frames. By aggregating data from multiple frames and calculating statistical metrics (mean, standard deviation) across the group, the system achieves more accurate roll angle calibration while managing processing complexity through structured combination
2Measurement precision
If multiple image frames are used for roll angle calibration, then the calibration accuracy is improved, but the processing time increases
Solution Approach 1:
The patent applies partial action by processing only a selected number of consecutive frames in groups rather than all available frames. This selective approach provides sufficient statistical data for accurate roll angle calibration while avoiding the excessive processing time that would result from analyzing every frame, achieving an optimal balance between accuracy and efficiency
3Manufacturing precision
If lane line fitting is performed on single frame, then the computational load is low, but the fitting error increases
Solution Approach 1:
The patent transitions from two-dimensional single-frame analysis to a multi-dimensional approach by incorporating temporal information across multiple consecutive frames. This dimensional expansion allows the system to reduce fitting errors through statistical aggregation while managing computational power through structured group processing and efficient statistical calculations
Data Source
AI summary
This application relates to the technical field of Internet of Vehicles, and discloses a method and an apparatus for calibrating a roll angle of an on-board camera, a device, and a storage medium. The method for calibrating a roll angle of an on-board camera includes obtaining an image frame sequence collected by the on-board camera while a vehicle is in motion, the image frame sequence comprising n image frames of a road captured by the on-board camera at n timestamps; selecting a plurality of image groups from the image frame sequence; obtaining lane line images corresponding to the plurality of image groups at a target roll angle, to obtain a lane line image set corresponding to the target roll angle; and determining the roll angle of the on-board camera based on lane line image sets corresponding to different roll angles.