Vehicle Camera Calibration Using Reference Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing camera calibration methods for vehicles require precise and time-consuming measurements, often necessitating immobilization on a test bench, which limits flexibility and efficiency, especially for cameras used in driver assistance systems.
Innovation Solution
A method and device for calibrating cameras on board vehicles using a reference sensor to determine calibration parameters by matching actual object positions with their image positions in shots, allowing calibration without immobilization, using a rotation and translation matrix to switch between reference frames, and iteratively minimizing differences between theoretical and actual image positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing calibration procedures are used, then measurement precision is improved, but device complexity and time consumption increase
Solution Approach 1:
The patent uses a reference sensor to capture actual object positions and creates a correspondence between these real-world positions and camera image positions. This copying approach replaces complex physical measurement instrumentation with data correspondence establishment, achieving accurate calibration parameters without requiring precision measurement devices.
Solution Approach 2:
The patent replaces mechanical measurement systems (test benches, precision instrumentation for angle and shift measurements) with a computational approach using reference sensor data and image processing. The mechanical calibration process is substituted by algorithms that compute calibration parameters from position correspondences.
2Measurement precision
If existing calibration procedures are used, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent performs preliminary data collection by the reference sensor to capture actual object positions before camera calibration. This preliminary action prepares the necessary correspondence data in advance, allowing the calibration computation to proceed efficiently without time-consuming measurements during the actual calibration process.
Solution Approach 2:
By copying actual object positions from the reference sensor and matching them with camera image positions, the patent eliminates time-consuming measurement steps. The copying process directly provides the calibration data needed, significantly reducing calibration time while maintaining precision.
3Manufacturing precision
If existing calibration procedures are used, then calibration accuracy is improved, but ease of operation deteriorates
Solution Approach 1:
The reference sensor on the vehicle performs self-service by automatically capturing actual object positions without requiring external measurement equipment or complex setup. The system uses its own existing sensor to provide calibration data, making the process simple and self-contained.
Solution Approach 2:
The reference sensor serves multiple functions: it acts as both a operational sensor for vehicle functions and a calibration reference source. This multi-functionality eliminates the need for separate calibration equipment, simplifying the calibration operation while maintaining accuracy.
4Measurement precision
If vehicle immobilization on test bench is required, then measurement precision is improved, but productivity decreases
Solution Approach 1:
The patent enables dynamic calibration by allowing the vehicle to move during the calibration process. Instead of requiring static immobilization on a test bench, the system captures calibration data while the vehicle is in motion, significantly improving productivity without sacrificing measurement precision through the use of reference sensor data.
Data Source
AI summary
A method for calibrating a camera on board a motor vehicle using a reference sensor on board the vehicle includes: a) acquiring, using the reference sensor, actual positions of at least one object in the vehicle surroundings, b) taking, using the camera, a shot each time one of the actual positions is acquired by the reference sensor, c) determining the position of the image of each object in the shots taken by the camera, d) forming position pairs by matching each actual position of each object with the position of the image of the object in the shot taken by the camera at the time of acquiring the actual position of the object, e) determining, using a computing unit, parameters for calibrating the camera from the position pairs formed.


