Raster to Vector Character Conversion via Adaptive Approximation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Converting scanned document images into vector format increases data volume significantly, especially for characters with complex shapes, leading to potential shape alteration when reducing contour points.
Innovation Solution
An image processing apparatus that extracts characters from raster data, determines whether to use curve or straight-line approximation based on character size, using straight-line approximation for smaller characters and curve approximation for larger ones to maintain shape integrity while reducing data volume.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If curve approximation is used to convert raster images to vector format, then character shape accuracy is improved, but data amount increases significantly
Solution Approach 1:
The patent applies different approximation methods (straight line vs. curve) to different character regions based on local characteristics. Complex curved regions use curve approximation to maintain shape accuracy, while simple regions use straight line approximation to reduce data amount. This local differentiation resolves the contradiction between shape accuracy and data volume.
Solution Approach 2:
The patent dynamically adjusts the approximation method parameter based on character size and shape characteristics. When character size exceeds a threshold or curvature is detected, curve approximation is activated; otherwise, straight line approximation is used. This parameter change strategy optimizes the balance between shape fidelity and data efficiency.
2Manufacturing precision
If the number of contour points is increased to maintain character shape, then character shape accuracy is improved, but data amount increases
Solution Approach 1:
The patent uses curvature detection to identify regions where curve approximation is necessary. Instead of uniformly increasing contour points throughout the character, the system applies curve approximation only to regions with significant curvature, thereby maintaining shape accuracy where needed while minimizing data usage in straight regions.
Solution Approach 2:
The patent differentiates the treatment of different character regions based on their geometric properties. Regions with high curvature receive enhanced approximation with more contour points, while regions with low curvature use simpler straight line approximation. This local quality differentiation maintains shape accuracy without uniformly inflating data volume.
3Device complexity
If conventional outlining is applied to all characters uniformly, then processing simplicity is maintained, but data amount increases unnecessarily for simple characters
Solution Approach 1:
The patent implements a dynamic decision-making process that adapts the approximation method based on real-time character analysis. The system evaluates character size and shape characteristics for each character, then dynamically selects the appropriate approximation method. This dynamic approach maintains processing simplicity through automated decision-making while avoiding unnecessary data generation for simple characters.
Solution Approach 2:
The patent changes the processing parameter (approximation method) based on character characteristics. When character size is below a threshold or the character appears to have simple geometry, the system switches to straight line approximation, thereby reducing data amount while maintaining processing simplicity through parameter adaptation rather than complex uniform processing.
Data Source
AI summary
A labeling unit that extracts a character included in raster format image data and finds the size of the extracted character, and an approximation method determination unit that determines whether or not to use curve approximation to convert the character based on the found size, are provided. In the case where it has been determined to use curve approximation, the character is converted into a vector format by performing straight-line approximation or curve approximation on each of the various parts of the contour of the character, whereas in the case where it has been determined not to use curve approximation, the character is converted into a vector format by performing straight-line approximation and not performing curve approximation on each of the various parts of the contour of the character.