Image Color Conversion Using Sparse 3D-LUT and Gamma Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing techniques for image color conversion using three-dimensional look-up tables (3D-LUTs) face challenges with large data sizes due to high lattice point density, which affects accuracy and processing efficiency, especially when dealing with 24-bit color images.
Innovation Solution
The proposed solution involves generating gamma correction processing information based on achromatic color input values and referencing a 3D-LUT to acquire representative output values, which are then subjected to gamma correction and color difference signal processing, allowing for accurate color conversion without increasing the number of lattice points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the density of discrete lattice points in a three-dimensional LUT is increased to improve color conversion accuracy, then the data size becomes significantly larger
Solution Approach 1:
The invention segments the color conversion process into two distinct stages: a first color conversion using a three-dimensional LUT with discrete lattice points, and a second color conversion using interpolation calculation between those lattice points. This segmentation allows the system to use a sparse LUT (reducing data size) while achieving high accuracy through the combined two-stage process, effectively resolving the contradiction between data size and conversion accuracy.
2Measurement precision
If a three-dimensional LUT is used for all gray levels to achieve accurate color conversion, then the number of lattice points becomes approximately 17 million, significantly increasing data size
Solution Approach 1:
The invention extracts only the essential lattice points needed for color conversion from the complete three-dimensional color space, rather than using all possible gray level combinations. By selecting discrete lattice points strategically and using them as reference points for interpolation, the system achieves accurate color conversion with a much smaller set of stored values, reducing both data size and structural complexity.
3Quantity of substance
If interpolation calculation is performed among discrete lattice points to reduce data size, then processing complexity increases
Solution Approach 1:
The invention performs preliminary organization of the discrete lattice points in the three-dimensional LUT, arranging them in a structured format that facilitates efficient interpolation calculations. By pre-organizing the lattice point data and establishing a systematic interpolation framework, the system reduces the computational complexity of the interpolation process while maintaining data size efficiency.
Data Source
AI summary
At least one processor configured as hardware generates gamma correction processing information based on an achromatic color input value and an output value acquired by referencing a three-dimensional LUT based on the achromatic color input value, references the three-dimensional LUT based on a representative input value as a chromatic color input value to acquire a representative output value, sets color difference signal processing information such that the representative output value is obtained when the representative input value is subjected to gamma correction processing and color difference signal processing, and subjects an image signal to gamma correction processing and color difference signal processing.