Image Processing Device Calibration Using 3DLUT and 1DLUT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for achieving high accuracy in color reproduction, such as those using three-dimensional lookup tables (3DLUTs), result in enlarged circuit scales, making it difficult to realize accurate color reproduction with a small circuit size.
Innovation Solution
A method involving a combination of three-dimensional and one-dimensional lookup tables, where initial processing adjusts for target color temperature and space, followed by fine adjustments, and correction based on colorimetric measurements, allows for accurate color reproduction even with a reduced number of grids in the 3DLUT, using calibration information and gain adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of grids in the three-dimensional lookup table is increased to achieve higher accuracy in color reproduction, then the accuracy in color reproduction is improved, but the circuit scale becomes substantially larger
Solution Approach 1:
The patent segments the color reproduction processing into two distinct lookup tables: a three-dimensional lookup table for color space conversion and a one-dimensional lookup table for color temperature adjustment. This segmentation allows each table to have a reduced number of grids while collectively achieving high accuracy, thereby reducing the overall circuit scale compared to a single large 3DLUT.
Solution Approach 2:
The one-dimensional lookup table acts as an intermediary component that processes the output of the three-dimensional lookup table. By introducing this intermediate processing stage, the system can achieve fine-tuned color temperature control without requiring an excessive number of grids in the primary 3DLUT, thus maintaining small circuit scale while improving accuracy.
2Device complexity
If the number of grids in the three-dimensional lookup table is reduced to decrease circuit scale, then the circuit scale becomes smaller, but the accuracy in color reproduction deteriorates
Solution Approach 1:
By dividing the color processing function into two specialized lookup tables with fewer grids each, the system maintains or improves color reproduction accuracy despite reducing the circuit scale. The 3DLUT handles color space conversion while the 1DLUT handles color temperature adjustment, allowing both to be optimized with smaller grid counts.
Solution Approach 2:
The patent applies different processing characteristics to different parts of the color reproduction pipeline: the 3DLUT is optimized for color space conversion while the 1DLUT is optimized for color temperature adjustment. This local specialization allows each component to achieve high accuracy with fewer grids, improving overall efficiency and reducing circuit scale.
3Manufacturing precision
If adjustment for target color space is performed before adjustment for target color temperature, then the color space mapping is accurate, but color temperature differences may remain that lower color reproduction accuracy
Solution Approach 1:
The system performs color space conversion as a preliminary action using the 3DLUT, then follows up with color temperature adjustment using the 1DLUT. This sequential approach ensures that color space mapping is established first, then refined with precise color temperature control, achieving both objectives with high accuracy.
Solution Approach 2:
The patent implements a feedback mechanism where the color temperature adjustment in the second processing stage corrects any color temperature deviations that may have been introduced during the color space conversion in the first stage. This feedback loop ensures that both color space accuracy and color temperature accuracy are achieved.
Data Source
AI summary
A method for calibrating an image processing device performs first processing to first image data, using a three-dimensional lookup table, and performing second processing to image data that has been processed by the first processing, using a one-dimensional lookup table on the output side of the three-dimensional lookup table. The method includes first setting step for setting first calibration information for each color component to have a target color temperature at the one-dimensional lookup table, second setting step for setting second calibration information for mapping the first image data to a target color space at the three-dimensional lookup table, and correction step for correcting the first calibration information based on an error from a target characteristic using image data that has been processed by the second processing.


