In-Vehicle Camera Calibration Using Road Feature Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing calibration methods for in-vehicle cameras are limited in their ability to accurately calibrate external parameters while the vehicle is in motion, leading to potential errors in distance calculations due to changes in vehicle posture and external conditions.
Innovation Solution
A calibration apparatus and method that includes an image acquisition unit, feature point extraction unit, tracking unit, lane recognition unit, sorting unit, and external parameter estimation unit, which acquires and processes images to track feature points, recognize the vehicle's lane, sort valid feature point trajectories, and estimate external camera parameters, allowing for accurate calibration during vehicle operation without additional sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If calibration is performed using a specified test pattern in the factory before shipment, then the external parameters can be calculated in advance, but the calibration accuracy deteriorates when vehicle posture changes due to passenger load or cargo movements
Solution Approach 1:
The patent transitions from static factory calibration to dynamic on-road calibration. The system continuously captures images while the vehicle is moving, tracks feature points across multiple frames, and estimates external parameters in real-time based on actual vehicle posture conditions, making the calibration adaptive to changing vehicle states
Solution Approach 2:
The calibration system uses the vehicle's own camera and naturally occurring road features (lane markings) as the calibration target. The vehicle calibrates itself using its existing sensing capabilities without requiring external calibration equipment or stopping the vehicle, transforming the calibration process into a self-service operation
2Adaptability or versatility
If calibration is performed while the vehicle is running on the road, then the adaptability to vehicle posture changes is improved, but the situation capabilities for calibration are limited and measurement precision deteriorates
Solution Approach 1:
The patent extracts and isolates feature points that lie specifically on the road surface plane from the overall image data. By separating road surface feature points from other features (buildings, vehicles, sky), the system ensures that only relevant, planar features are used for calibration, improving measurement precision
Solution Approach 2:
The patent introduces lane marking recognition as an intermediary step between raw image capture and external parameter estimation. The lane recognition unit identifies the vehicle's driving lane, and this information serves as a mediator to guide the selection and validation of feature points, ensuring they are appropriate for calibration purposes
3Productivity
If all feature point trajectories are used for external parameter estimation, then the processing speed is improved, but the manufacturing precision deteriorates due to inclusion of invalid trajectories
Solution Approach 1:
The patent maintains continuous calibration processing by continuously capturing images, tracking feature points across frames, and constantly updating external parameter estimates. This continuous process ensures that valid feature point trajectories are consistently utilized without interruption, maintaining both processing efficiency and precision
Solution Approach 2:
The patent replaces manual or offline mechanical calibration processes with automated computer vision-based tracking and estimation. The system automatically identifies, tracks, and validates feature points through image processing algorithms, substituting mechanical calibration equipment with software-based intelligent processing
Data Source
AI summary
Calibration with high accuracy can be realized even when performing the calibration while running on the actual road. Specifically, the calibration apparatus is mounted in a vehicle and includes: an image acquisition unit configured to acquire captured images obtained by a camera, which is mounted in the vehicle, capturing images of surroundings of the vehicle; a feature point extraction unit configured to extract a plurality of feature points from the captured images; a tracking unit configured to track the same feature point from a plurality of the captured images captured at different times with respect to each of the plurality of feature points, which are extracted by the feature point extraction unit, and record the tracked feature point as a feature point trajectory; a lane recognition unit configured to recognize an own vehicle's lane which is a driving lane on which the vehicle is running, from the captured images; a sorting unit configured to sort out the feature point trajectory, which is in the same plane as a plane included in the own vehicle's lane recognized by the lane recognition unit, among feature point trajectories tracked and recorded by the tracking unit; and an external parameter estimation unit configured to estimate external parameters for the camera by using the feature point trajectory sorted out by the sorting unit.