Pseudo-skeleton Image Enhancement for Document OCR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing document image enhancement methods for printing and imaging processes tend to make text strokes thicker, negatively impacting text features and performance in Optical Character Recognition (OCR) and document authentication systems.
Innovation Solution
A method that segments document images into non-graphical and graphical regions, binarizes text regions, computes multiple levels of pseudo-skeletons, and applies inverse-degradation functions to pixel values in pseudo-skeleton difference areas to enhance text strokes while maintaining original thickness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If edge enhancement or contrast mapping is applied to improve image quality, then image quality is improved, but text stroke thickness increases
Solution Approach 1:
The image is segmented into text regions and non-text regions. Different enhancement strategies are applied to each region: pseudo-skeleton based enhancement for text regions to preserve stroke thickness, and traditional edge enhancement for non-text regions. This segmentation allows simultaneous optimization of image quality and text characteristics.
Solution Approach 2:
Different enhancement methods are applied to different parts of the image based on their specific requirements. Text regions use pseudo-skeleton based enhancement that preserves local stroke thickness, while non-text regions can use more aggressive edge enhancement. This local differentiation resolves the contradiction by optimizing each region according to its specific needs.
2Illumination intensity
If text strokes are enhanced to improve visual conspicuousness, then visual conspicuousness is improved, but text features for OCR and authentication deteriorate
Solution Approach 1:
Before applying enhancement, the text regions are identified and segmented. Pseudo-skeletons are computed to establish the original stroke structure. This preliminary action allows the enhancement process to reference the original stroke geometry and preserve it during enhancement, preventing degradation of text features needed for OCR and authentication.
Solution Approach 2:
The pseudo-skeleton represents a copy or abstraction of the original text stroke structure. By using this skeleton copy as a reference during enhancement, the system can restore and enhance visual appearance while maintaining fidelity to the original stroke geometry, thus preserving text features for measurement and recognition.
Data Source
AI summary
In a document image enhancement method, text character strokes are enhanced in a way such that areas closer to the center (skeleton) of each stroke are enhanced (e.g. made darker) by greater amounts than areas farther away from the center. Each text or line region of the input image is first binarized to generate a binary image containing connected components corresponding to character strokes and lines. Multiple levels of pseudo-skeletons are computed for each connected component, for example by using successive thinning. Multiple pseudo-skeleton difference areas, i.e. differences between successive levels of pseudo-skeletons, are generated. Pixels located in different pseudo-skeleton difference areas are enhanced by different amounts, by applying different inverse-degradation functions. Graphical regions of the input image may be treated with edge enhancement.