Variable Glyph Encoding for Text Data Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data protection methods fail to effectively prevent unauthorized extraction of text data from documents, as standard encoding schemes like ASCII or Unicode allow straightforward processing and copying of character codes.
Innovation Solution
Converting textual data into explicit drawing instructions for each character, combining multiple glyphs into single instructions, and applying geometric transformations to obscure the connection between original text and drawing commands, using SVG standards and mechanisms to hinder text extraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard encoding schemes like ASCII or Unicode are used to store text characters, then the document can be easily processed and text can be extracted by copying character codes, but data protection is insufficient and unauthorized interpretation is possible
Solution Approach 1:
The patent changes the encoding parameters by replacing fixed character codes with variable-length drawing instructions. Each character is represented by a unique sequence of drawing commands that render its visual form, making the encoding scheme adaptable and resistant to standard text extraction methods while maintaining rendering capability
Solution Approach 2:
The patent introduces drawing instructions as an intermediary layer between the original text and its visual representation. Instead of directly storing or transmitting character codes, the system uses drawing commands as a mediator that preserves the visual appearance while obscuring the underlying text content, thereby enhancing data protection
2Reliability
If text is stored using uniform encoding, then processing and extraction are straightforward, but unauthorized copying and interpretation are easy
Solution Approach 1:
The patent inverts the traditional approach by not directly encoding text characters but instead encoding the visual representation of characters through drawing instructions. This inversion makes text extraction difficult because the drawing commands must be interpreted and reversed to retrieve the original text, thereby enhancing data protection while maintaining visual fidelity
3Reliability
If drawing instructions are used to represent each character, then text extraction is hindered, but the document still renders correctly
Solution Approach 1:
The patent segments each character into multiple drawing instructions that collectively render the character's visual form. By dividing the representation of each character into discrete drawing commands, the system obscures the original text while preserving the visual output, thereby enhancing data protection without compromising rendering capability
Data Source
AI summary
A system and method for graphically encoding text. A textual data set comprising a plurality of encoded text characters that corresponding to at least one text character to be visually rendered is received. Based on the textual data set, a drawing instruction set is determined that has a respective drawing instruction to draw at least part of a glyph of each of the encoded text characters. Each at least one drawing instruction excludes an indication of a correspondence with an associated text character code, and each respective at least one drawing instruction for a specified text character excludes instructions to repeat other drawing instructions for other instances of the specified text character.