Automatic White Balance Calibration Using Dynamic Golden Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic white balance (AWB) calibration methods fail to accurately calibrate devices with larger manufacturing errors, leading to poor color temperature estimations and reduced imaging quality, as they result in calibration gains that are too large or too small, causing discrepancies between the white points of the calibrated device and the golden set.
Innovation Solution
The method adjusts the red and blue deviation values of the golden set when the differences between the device under calibration and the golden set exceed predetermined thresholds, reducing these differences to achieve more accurate calibration gains, ensuring the device aligns with the white area under various color temperatures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If current AWB calibration methods are used, then devices with small manufacturing errors can be calibrated accurately, but devices with large manufacturing errors produce calibration gains that are too large or too small, causing poor calibration outcomes
Solution Approach 1:
The patent adjusts the golden values locally based on the device under test's characteristics. Instead of using fixed golden values for all devices, the system dynamically determines appropriate golden values by comparing the DUT's white point with the standard white point, and adjusting the golden values accordingly. This allows the calibration process to adapt to devices with different error magnitudes while maintaining calibration accuracy.
Solution Approach 2:
The patent changes the parameters used in calibration by introducing adjustable golden values that can be modified based on the device's error characteristics. The system calculates the difference between the DUT's white point and the standard white point, then adjusts the golden values (R_golden, B_golden) to compensate for manufacturing errors. This parameter adjustment enables accurate calibration of devices with large errors without producing excessively large or small calibration gains.
2Ease of manufacture
If fixed golden values are used for calibration, then the calibration process is simple, but devices with large manufacturing errors cannot be accurately calibrated
Solution Approach 1:
The patent performs preliminary assessment of the device under test before final calibration. The system first captures the DUT's white point under calibration light, compares it with the standard white point, and determines the magnitude of manufacturing errors. Based on this preliminary assessment, the system then selects or adjusts appropriate golden values for the calibration process. This preliminary action ensures that the calibration process remains straightforward while adapting to different device characteristics.
Solution Approach 2:
The patent introduces dynamic adjustment of golden values during the calibration process. Instead of using static, fixed golden values, the system dynamically determines the appropriate golden values based on the DUT's actual white point characteristics. The golden values are adjusted in real-time based on the measured differences between the DUT and standard white points, enabling the calibration process to adapt to varying device conditions while maintaining simplicity.
Data Source
AI summary
A method, apparatus, computer-readable storage medium, and terminal for calibrating automatic white balance is provided including: obtaining a first white point acquired by a device under calibration capturing a neutral color under a calibration light source, and obtaining a second white point acquired by a predetermined golden set capturing a neutral color under the calibration light source; if either or both a first difference value between a first red deviation value and a second red deviation value is greater than a first predetermined threshold value and a second difference value between a first blue deviation value and a second blue deviation value is greater than a second predetermined threshold value, performing one or more of adjusting the second red deviation value so that the first difference value becomes smaller and adjusting the second blue deviation value so that the second difference value becomes smaller; and calibrating the device under calibration by using one or more of the adjusted second red deviation value and the adjusted second blue deviation value as a first golden value and a second golden value, respectively.