Image Processing Apparatus for Scanned Illustration Blurring Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image processing techniques struggle with blurring in boundary parts of illustrations during scanning, leading to increased vector data size and degraded image quality, especially when fine line art regions are vectorized with a lower limit line width, causing lines to appear thicker and blurred.
Innovation Solution
An image processing apparatus and method that divides input images into color regions, labels them, analyzes shape features, determines micro-regions, and combines them with neighboring regions to reduce blurring and improve vectorization accuracy, thereby reducing data size and enhancing reusability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If clustering is applied to reduce scan noise and extract outlines, then color separation accuracy is improved, but blurring occurs in fine line art regions and edge components
Solution Approach 1:
The patent segments the image processing into distinct stages: initial clustering for color separation, followed by micro-region detection and elimination. This segmentation allows the system to first achieve color separation accuracy through clustering, then separately address the blurring issue by identifying and eliminating micro-regions caused by scan noise, rather than trying to solve both problems simultaneously
Solution Approach 2:
The patent extracts and eliminates micro-regions (small clustered regions caused by scan noise) from the image after initial clustering. By specifically identifying and removing these problematic micro-regions through the micro-region elimination processing, the system removes the source of blurring while preserving the color separation benefits of clustering
2Shape
If vectorization is performed on fine line art regions, then illustration detail is preserved, but line width becomes thicker and blurred due to renderer lower limit
Solution Approach 1:
The patent performs micro-region elimination processing before vectorization to prevent the generation of problematic vector data. By eliminating micro-regions in advance, the system prevents the creation of overly detailed vector data that would cause rendering issues, thereby maintaining line width accuracy in the final vectorized output
3Measurement precision
If color regions are separated and vectorized individually, then color accuracy is improved, but data size increases due to generation of unnecessary vector data in blurring regions
Solution Approach 1:
The patent extracts and eliminates micro-regions that generate unnecessary vector data. By removing these micro-regions before vectorization, the system prevents the creation of excessive vector data while preserving the color accuracy benefits of region-based processing
Solution Approach 2:
The patent discards micro-regions identified as noise-induced artifacts through the micro-region elimination process. By discarding these unnecessary regions before vectorization, the system reduces the total amount of vector data generated while maintaining the essential color and shape information
Data Source
AI summary
For the problem of blurring at a boundary in a scanned illustration, a method of separating and vectorizing color regions accurately is realized by identification of a labeled region corresponding to the blurring and combining processing with another neighboring region. An image input unit, a color region dividing unit dividing an input image into the color regions, a labeling unit providing an individual label to the color region, a labeled region shape analysis unit obtaining labeled region shape information including information representing a shape feature of a labeled region which is the color region provided with the label, a micro-region determination unit determining whether the labeled region provided with the label is a micro-region or not according to the labeled region shape information, and a micro-region combining unit combining the region determined to be the micro-region with a neighboring region, are comprised.