Camera Calibration Device Using Feature Point Trajectories

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing camera calibration methods require a white line on the road to calibrate internal parameters, which is not feasible in all scenarios, leading to errors in distance calculations due to lens distortion changes caused by temperature and heat expansion.

Innovation Solution

A camera calibration device that extracts and tracks feature points from images captured by a vehicle-mounted camera, estimates trajectories based on vehicle movement, and calculates internal parameters, including lens distortion coefficients, using polynomial approximation and correspondence determination to minimize trajectory differences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a white line on the road is used for calibration, then internal parameters can be estimated, but calibration cannot be performed without the white line

Engineering Contradiction:
Improveinternal parameter estimation accuracyVSAvoidcalibration applicability
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The invention extracts the calibration function from the specific context of white line detection and generalizes it to work with any feature points on the road surface. By separating the calibration logic from the white line constraint, the system can now use diverse road features (markings, edges, objects) for calibration, thereby improving adaptability while maintaining measurement precision

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The calibration device is designed to handle multiple types of road features universally. Instead of being specialized for white lines only, the system can process various feature points including road markings, curbs, and other road surface elements, making the calibration function versatile and applicable in diverse driving scenarios

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Productivity

If internal parameters are derived in advance, then production efficiency is improved, but lens distortion changes due to temperature and heat expansion cause errors in distance calculation

Engineering Contradiction:
Improveproduction efficiencyVSAvoiddistance calculation accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system transitions from static pre-derived internal parameters to dynamic calibration that adapts to changing conditions. By performing calibration during vehicle movement using real-time feature point tracking, the system adjusts internal parameters to account for thermal expansion and distortion changes, maintaining distance calculation accuracy while preserving production efficiency through automated calibration

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The calibration system uses feedback from tracked feature point trajectories to continuously refine internal parameters. By comparing expected trajectories with actual observed trajectories and adjusting parameters accordingly, the system compensates for distortion changes caused by temperature variations, ensuring accurate distance calculations despite environmental conditions

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3416367B1Camera calibration device
Publication Date: 2020.08.26 FAURECIA CLARION ELECTRONICS CO LTD
  • EP3416367B1 patent drawingFigure 1
  • EP3416367B1 patent drawingFigure 2
  • EP3416367B1 patent drawingFigure 3

AI summary

This camera calibration device is provided with: an image acquisition unit into which a plurality of images captured at different times by a camera mounted to a vehicle are inputted; a feature point extraction unit which extracts features points from each of the images inputted into the image acquisition unit; a feature point tracking unit which tracks the feature points between the plurality of images; a vehicle movement acquisition unit which acquires information related to the movement of the vehicle; a trajectory estimation unit which, on the basis of the information acquired by the vehicle movement acquisition unit, estimates the trajectory of the feature points corresponding to the movement of the vehicle; and an internal parameter estimation unit which estimates internal parameters of the camera on the basis of the trajectory of the feature points tracked by the feature point tracking unit, and the trajectory of the feature points estimated by the trajectory estimation unit.