Embroidery Data Creation with Region-Specific Color Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing embroidery data creation technologies face challenges in naturally representing images with a wide range of colors, particularly in areas like human skin tones, as they often result in unnatural color mixing when reducing colors for embroidery threads.
Innovation Solution
A device and method that acquire thread color data, calculate the ratio of specific objects within an image, and allocate specific thread colors to corresponding areas, ensuring accurate color representation by selecting thread colors within defined threshold ranges to create embroidery data that mimics the original image effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If color reduction is performed to reduce the number of thread colors, then the complexity of embroidery data is reduced, but the natural appearance of the image is degraded
Solution Approach 1:
The patent applies local quality by differentiating treatment for different regions of the image. Specifically, it identifies regions containing specific objects (such as human faces) and applies stricter color preservation rules to those regions while allowing more aggressive color reduction in other areas. This ensures that critical areas maintain natural appearance while overall complexity is reduced.
Solution Approach 2:
The patent segments the image into different regions based on object detection and color characteristics. It divides the image into areas containing specific objects and areas without such objects, then applies different color allocation strategies to each segment. This segmentation allows optimized color representation in critical regions while simplifying non-critical regions.
2Ease of manufacture
If thread colors are selected from a limited set, then the feasibility of embroidery production is improved, but the accuracy of color representation deteriorates
Solution Approach 1:
The patent performs preliminary actions by pre-identifying regions containing specific objects and pre-determining appropriate thread color ranges for those regions before the main color allocation process. This allows the system to reserve suitable colors in advance for critical areas, ensuring accurate color representation is maintained while working within the constraints of limited thread color sets.
Solution Approach 2:
The patent changes parameters dynamically based on region characteristics. It adjusts the color selection criteria and allocation strategies according to whether a region contains specific objects or not. This parameter adaptation allows the system to optimize both manufacturability and color accuracy by adjusting the balance between limited color palette and representation fidelity.
3Adaptability or versatility
If mixed color expression is used to represent multiple colors, then the versatility of color representation is improved, but the natural appearance of specific objects deteriorates
Solution Approach 1:
The patent applies local quality by making the color representation strategy location-dependent. In regions containing specific objects like human faces, it uses single-color or limited-color assignments to maintain natural appearance. In other regions, it employs mixed color expression to achieve versatility. This localized differentiation resolves the contradiction between versatility and natural appearance.
Solution Approach 2:
The patent segments the color representation approach based on image content analysis. It separates the image into regions with specific objects and regions without, then applies appropriate color strategies to each segment. This segmentation enables mixed color expression in non-critical areas while maintaining natural appearance in critical areas.
Data Source
AI summary
A non-transitory computer-readable medium storing computer-readable instructions that, when executed by a processor of a device, cause the device to acquire pieces of thread color data, acquire image data representing an image, arrange line segments based on the image data, calculate a ratio of a first area occupied by a specific object with respect to the image, identify one or more pieces of first thread color data based on the ratio, identify one or more pieces of second thread color data based on the image data, allocate, to one or more of first line segments corresponding to the first area, first specific thread color data among the first thread color data, allocate, to one or more of second line segments corresponding to the second area, second specific thread color data among the first thread color data and the second thread color data, connect the line segments, and create embroidery data.