Auto Calibration for OIS Camera Gyroscope Sensitivity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for calibrating optical image stabilizer (OIS) cameras are inefficient, requiring lengthy processing times and increased costs to ensure exact performance due to variations in manufacturing states of gyroscope and actuator components.
Innovation Solution
An auto calibration method that involves removing DC offset, applying vibrator signals to measure gyroscope sensitivity, calculating gain values for both gyroscope and actuator, and measuring pixel displacement to adjust sensitivity, thereby reducing processing time and ensuring high performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual searching for gain value is used to calibrate OIS components, then processing time is decreased, but exact OIS performance cannot be ensured when component distribution is large
Solution Approach 1:
The calibration system performs automatic self-calibration by capturing images at multiple gain values, evaluating image quality metrics automatically, and selecting the optimal gain value without manual intervention. This eliminates the need for manual searching while ensuring exact OIS performance through systematic evaluation.
Solution Approach 2:
The system implements feedback by capturing test images, evaluating image quality metrics (such as sharpness, contrast, or noise levels), and using this evaluation to automatically adjust and select the optimal gain value. This closed-loop feedback mechanism ensures both speed and accuracy in calibration.
2Manufacturing precision
If automatic searching for gain value through image test is used, then exact OIS performance can be ensured, but processing time is considerably increased and processing cost is increased
Solution Approach 1:
The system performs a limited number of image captures at strategically selected gain values rather than exhaustive testing. By capturing images at a small set of representative gain values and using automated evaluation, the system achieves sufficient calibration accuracy without excessive processing time or cost.
3Manufacturing precision
If automatic searching for gain value through image test is used, then exact OIS performance can be ensured, but evaluation method becomes complicated and processing cost is increased
Solution Approach 1:
The system replaces complex manual evaluation methods with automated image processing algorithms. The controller automatically captures images, computes quality metrics, and determines optimal gain values, eliminating the need for complicated manual evaluation procedures and reducing overall system complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method significantly reduces processing time and cost while ensuring high performance by accurately calibrating the OIS camera's sensitivity, enabling efficient mass production.
Implementation Method 1
the gyroscope sensor generally has a function of sensing degrees of tilting with respect to two axes (pitch and yaw)
Implementation Method 2
applying a vibrator signal by operating a vibrator
Data Source
AI summary
An auto calibration method and an optical image stabilizer (OIS) camera using the same are provided. The auto calibration method includes removing a DC offset of a gyroscope. A vibrator signal applied by operating a vibrator, and an actual measurement value of the gyroscope is obtained using the applied vibrator signal. A first gain value for compensating for sensitivity of the gyroscope is calculated using the vibrator signal and the actual measurement value of the gyroscope, and an actuator is operated. A displacement of a pixel actually moving on an image is moved under an operation of the actuator. A second gain value for controlling a sensitivity variation of the actuator is calculated based on the displacement of the actually moving pixel. Accordingly, it is possible to reduce processing time and to ensure high performance.