Optimized Transformation Matrix for Visible Color Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing color space conversion methods, such as those using standard transformation matrices, inefficiently encode colors outside the visible gamut, leading to increased computational power and bandwidth usage without improving image quality, and result in higher quantization errors due to the representation of non-visible colors.
Innovation Solution
An optimized 3x3 transformation matrix is generated to convert XYZ color data into YCC data, focusing on encoding only visible colors by determining which color codes fall within the spectral locus and using a scaler function to adjust the transformation matrix, thereby optimizing the encoding process for visible colors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard transformation matrices are used for color space conversion, then all colors (including non-visible ones) are encoded, but this increases computational power and bandwidth usage without improving image quality
Solution Approach 1:
The patent extracts and removes the encoding of non-visible colors from the color space conversion process. By determining which color codes fall outside the spectral locus (non-visible colors) and excluding them from encoding, the system eliminates wasted computational resources while maintaining encoding of only the visible color spectrum that contributes to image quality
Solution Approach 2:
The patent modifies the transformation matrix parameters by applying a scaler function that adjusts the matrix based on the spectral locus boundaries. This parameter change optimizes the encoding process to map colors efficiently within the visible gamut, reducing the computational burden of encoding colors that cannot be perceived by the human eye
2Measurement precision
If standard transformation matrices are used for color space conversion, then all colors are represented, but this results in higher quantization errors due to the representation of non-visible colors
Solution Approach 1:
The patent extracts non-visible colors from the color space representation by identifying color codes outside the spectral locus and excluding them from the transformation process. This extraction eliminates the source of quantization errors that arise from attempting to represent colors that cannot be perceived, thereby improving the precision of the encoded color data
Solution Approach 2:
The patent changes the transformation matrix parameters through scaler optimization to better match the visible color gamut boundaries. This parameter adjustment ensures that quantization levels are optimally distributed across only the visible color spectrum, reducing quantization errors in the encoded image data
3Manufacturing precision
If an optimized transformation matrix is generated to encode only visible colors, then quantization errors are reduced and image quality is improved, but the device complexity increases
Solution Approach 1:
The patent performs preliminary determination of the spectral locus and pre-calculates the optimized transformation matrix parameters before the actual color space conversion process. By preparing the optimized matrix in advance with known spectral boundaries, the system reduces the complexity of real-time calculations while maintaining improved image quality through accurate visible color encoding
Solution Approach 2:
The patent introduces the scaler function as an intermediary that bridges the standard transformation matrix and the optimized visible-color-only encoding. This intermediary component systematically adjusts the matrix parameters based on spectral locus data, making the optimization process more manageable and less complex by breaking it into structured steps
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Novel methods and systems for color space conversions are disclosed, relating to the optimization of a transformation matrix to convert between wide gamut opponent color spaces. The optimization may be based on whether the color values are in or out of gamut.