Image Creation Device Using Dual Background Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for capturing yarn and textile images struggle to separate object images from background images, especially in translucent areas, leading to mixing and difficulties in achieving realistic simulations with accurate color and texture representation.
Innovation Solution
The system captures object images twice using different background images and solves the resulting system of equations to isolate the object's color and opacity, allowing for the creation of high-quality images that separate the object from the background effectively, even in high-transparency regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a single background image is used for capturing yarn images, then the capture process is simple, but the background image mixes with translucent portions of the yarn causing color distortion
Solution Approach 1:
The patent segments the image capture process into two separate captures with different backgrounds (white and black). This segmentation allows the system to later separate the yarn's true color from the background influence by solving a system of equations, thereby resolving the contradiction between simple capture and accurate color representation.
Solution Approach 2:
The patent changes the background parameter (from single to dual backgrounds with different luminance characteristics) to enable mathematical separation of the yarn's intrinsic color properties from the background mixing effect. This parameter change allows accurate color extraction while maintaining a relatively simple capture process.
2Measurement precision
If manual opacity adjustments or stencils are used to separate object from background, then separation accuracy improves, but the process complexity and time increase
Solution Approach 1:
The patent replaces manual mechanical separation methods (stencils, opacity adjustments) with an automated mathematical solution. By substituting the mechanical separation process with a system of equations that can be solved computationally, the patent achieves high separation accuracy without increasing process complexity or requiring manual intervention.
3Manufacturing precision
If multiple captures with different backgrounds are performed, then background mixing is reduced, but the capture time and process complexity increase
Solution Approach 1:
The patent segments the image formation into two distinct captures with different backgrounds, which when combined with mathematical processing, achieves high-quality yarn images. The segmentation approach allows efficient processing where the system of equations can be solved rapidly, minimizing the time penalty of multiple captures.
Solution Approach 2:
By changing the background parameter between captures (white to black), the patent enables mathematical extraction of the yarn's true properties. This parameter change strategy achieves superior image quality while the computational solution keeps the overall process time acceptable.
Data Source
AI summary
A captured image of an object is input twice using a white background and a black background, after which the opacity D and the color image X of the object are obtained to yield an object image. It is possible to easily create a high-quality real image.