Synthesized Font Generation for Scanned Document Text Editing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Scanned documents lack typeface information, leading to a loss of fidelity when edited, as default fonts are used without regard to the original layout, resulting in a poor user experience.
Innovation Solution
A method and apparatus that create a synthesized font from scanned documents, allowing for text editing while maintaining the original appearance by replacing characters with similar ones from the synthesized font and adding new characters that match the original document's style, using optical character recognition and system fonts to ensure fidelity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If text editing is enabled in a scanned document using a default font, then text editing capability is provided, but the fidelity of the document's appearance is lost
Solution Approach 1:
The patent creates a synthesized font by copying and analyzing character images from the scanned document. The system extracts character data from the scanned image, generates a synthesized font that replicates the original typography, and uses this font for all subsequent text editing operations. This ensures that any text added or modified during editing maintains the same appearance as the original scanned text, resolving the contradiction between enabling editing and preserving appearance fidelity.
2Ease of operation
If a default font is used for text editing without regard to layout, then editing is simplified, but the layout consistency and professional appearance are compromised
Solution Approach 1:
The patent changes the fundamental parameter of font selection from a fixed default to a dynamically generated synthesized font. The system analyzes the scanned document's character images to create a customized font that matches the original layout, typography, and styling parameters. This synthesized font is then applied to all editing operations, ensuring layout consistency while maintaining ease of use through automated font generation rather than manual configuration.
3Device complexity
If scanned documents are treated as images without typeface information, then document processing is simplified, but text editing precision and appearance matching are impossible
Solution Approach 1:
The patent extracts character images and typographic information from the scanned document image, transforming the flat image data into structured font information. By taking out and analyzing individual character images, the system reconstructs the original typeface characteristics and creates a synthesized font. This process maintains the simplicity of processing scanned documents as images while enabling precise text appearance matching through the extracted and reconstructed font data.
Data Source
AI summary
A computer implemented method and apparatus for enabling text editing in a scanned document while maintaining fidelity of appearance of the text. The method comprises creating a synthesized font comprising a plurality of characters using characters present in a scanned document; replacing the plurality of characters in the scanned document with characters from the plurality of characters from the synthesized font; and enabling editing of the scanned document wherein enabling editing comprises adding at least some characters from the plurality of characters of the synthesized font to the document for at least some characters added during editing.