Non-Convex Gamut Mapping via Dual Boundary Descriptors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gamut mapping methods struggle with non-convex color gamuts, leading to overestimation of gamut volume and distortion of color differences, especially when dealing with devices having subtractive principles, as they hide the non-convex shape of the source gamut and complicate interface definitions between system modules.
Innovation Solution
A method that creates both convex and non-convex gamut boundary descriptors for source and target color devices, allowing pre-mapping, main-mapping, and post-mapping of colors to ensure accurate gamut mapping while allowing simple geometrical operations, using a plurality of Gamut Boundary Descriptions (GBDs) to handle non-convex gamuts effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If convex hull is used to represent gamut boundary, then geometrical operations become simple, but gamut volume is overestimated and color differences are distorted
Solution Approach 1:
The gamut boundary representation is segmented into two distinct descriptors: a convex hull for simple geometrical operations and a non-convex alpha shape for accurate color representation. This segmentation allows each descriptor to serve its specific purpose without compromise.
Solution Approach 2:
The patent introduces an intermediary non-convex alpha shape descriptor that bridges the gap between the simplified convex hull and the actual complex gamut boundary. This intermediary structure enables accurate color difference calculations while maintaining computational feasibility.
2Measurement precision
If non-convex gamut boundary is directly used, then color reproduction accuracy is improved, but system complexity and computational load increase
Solution Approach 1:
The gamut mapping system is segmented into distinct modules: convex hull generation, non-convex alpha shape generation, and conditional mapping logic. This segmentation manages complexity by organizing functions into separate, well-defined components.
Solution Approach 2:
Both convex and non-convex gamut descriptors are pre-computed and stored before actual color mapping operations. This preliminary action eliminates the need for complex real-time calculations during color conversion, reducing operational computational load.
3Ease of operation
If convex gamut is used to hide non-convex shape, then interface definition between modules is simplified, but gamut mapping precision deteriorates
Solution Approach 1:
The system implements a universal interface that accepts both convex and non-convex gamut descriptors. This multi-functionality allows the same interface to handle both simple and complex gamut cases without requiring separate handling logic.
Solution Approach 2:
The patent creates a copy of the gamut boundary in two forms: a convex approximation for simplified operations and a precise non-convex replica for accurate mapping. Both copies coexist and are used according to the specific operational requirements.
Data Source
Figure 1~2
Figure 3~6e
Figure 5
AI summary
Method of mapping source colors from a source color device having an actual source gamut that is represented by a source convex GBD (gamut boundary descriptor) and a source non-convex GBD into target colors for a target color device having an actual target gamut that is represented by a a target convex GBM and a target non-convex GBD, comprising, 1 - if the source color to map is located outside the source non-convex GBD, a pre-mapping step that maps the source color into the source non-convex GBD, 2 - if the mapped target color is located outside the target non-convex GBD, a post-mapping step that maps the target color into the target non-convex GBD. This method allows precise gamut mapping while allowing also for simple geometrical operations using convex gamuts.