Color Processing Device Composite LUT Inverse Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In color processing systems, multi-dimensional look-up tables (LUTs) used for color conversion and calibration perform linear interpolation, which can lead to imperfect color reproduction, and existing tone reproduction curves (TRCs) address this through non-linear tone modification but still result in suboptimal results when compositing multiple color conversion processes.
Innovation Solution
A color processing device that composites plural color conversion rules through serial processing and applies an inverse one-dimensional correlation to modify these rules, reducing the number of multi-dimensional LUTs required and improving color reproduction by ensuring non-linearity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple multi-dimensional LUTs are used for color conversion and calibration, then color conversion processes can be performed, but the number of processes increases and color reproduction accuracy deteriorates due to linear interpolation
Solution Approach 1:
The patent merges multiple multi-dimensional LUTs into a single composite LUT by compositing color conversion rules through serial processing. This consolidation reduces the number of separate color conversion processes while maintaining or improving color reproduction accuracy through the application of inverse one-dimensional correlations that introduce necessary non-linearity.
Solution Approach 2:
The patent changes the parameter representation by applying inverse one-dimensional correlations to transform the composited color conversion rules. This parameter transformation introduces non-linearity to correct the linear interpolation缺陷 of traditional multi-dimensional LUTs, thereby improving color reproduction accuracy without increasing process complexity.
2Manufacturing precision
If tone reproduction curves are applied for non-linear tone modification, then color reproduction can be improved, but the composition of multiple color conversion processes still results in suboptimal outcomes
Solution Approach 1:
The patent combines tone reproduction curve functionality with multi-dimensional LUT composition by applying inverse one-dimensional correlations during the compositing process. This integration eliminates the need for separate TRC application steps while achieving non-linear tone modification, thereby improving color reproduction without increasing device complexity.
Solution Approach 2:
The patent performs preliminary action by pre-computing and storing the composited color conversion rules with embedded inverse one-dimensional correlations in a single LUT structure. This preliminary compositing eliminates the need for sequential application of multiple conversion processes, reducing complexity while preserving color accuracy.
3Ease of operation
If linear interpolation is used in multi-dimensional LUTs, then color conversion is computationally simple, but color reproduction becomes imperfect
Solution Approach 1:
The patent transforms the linear interpolation parameter space by applying inverse one-dimensional correlations to the composited color conversion rules. This parameter change introduces non-linearity that corrects color reproduction errors while maintaining the single-LUT computational structure, thus preserving ease of operation while improving accuracy.
Data Source
AI summary
A color processing device includes a compositing unit, a modification unit, and an output unit. The compositing unit composites plural color conversion rules in a color space that includes plural color components through serial processing. The modification unit modifies the color conversion rules after being composited by the compositing unit, by applying an inverse conversion of a specific one-dimensional correlation, which correlates input color values and output color values in the color space, to the composited color conversion rules. The output unit outputs the composited color conversion rules after being modified and the specific one-dimensional correlation.