Character Identification via Glyph-Code Point Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Embedded font documents that lack code point data for certain characters cannot be correctly indexed or searched by computerized entities, despite appearing visually accurate to humans, due to the absence of necessary data for machine recognition.
Innovation Solution
Character identification data is created and stored, associating glyph data with code point data using hash functions or user input, enabling computers to recognize and identify characters in embedded font documents, even if they lack code point data, by mapping hash values to code point data for characters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If font subsetting is used to reduce document size by including only specific character font information, then the document size is reduced, but code point data may be omitted for certain characters
Solution Approach 1:
The system pre-generates a character identification data structure that maps glyph data to code point data before the actual need arises. This preliminary preparation ensures that when font subsetting occurs, the mapping information is already available to restore missing code point data, preventing information loss while maintaining document size reduction benefits
Solution Approach 2:
A character identification data structure acts as an intermediary between glyph data and code point data. When code point data is missing from subsetting, this intermediary structure provides the mapping information needed to recover the original character identity, enabling computers to 'read' characters without requiring the original code point data to be present
2Reliability
If code point data is embedded with every character to enable machine recognition, then computerized entities can correctly index and search documents, but the document size increases significantly
Solution Approach 1:
The invention extracts the character identification mapping information from the main document structure and stores it separately in a persistent data store. This extraction allows the document to reference character identities without embedding full code point data for every character, significantly reducing document size while maintaining machine recognition capability through the separate mapping structure
Solution Approach 2:
The solution segments the character identification system into two parts: (1) glyph data embedded in the document for visual rendering, and (2) code point data stored separately in a character identification data structure. This segmentation allows the document to remain compact while still enabling machine recognition through the separate data structure that maps glyphs to their corresponding code points
Data Source
AI summary
Approaches for enabling a computerized entity to recognize characters in an electronic document. In a persistent data store, character identification data is stored. Character identification data is data that, for one or more characters of one or more fonts, associates (a) glyph data for a character with (b) code point data for the character, where the glyph data describes how to render the character on or to an output device, and the code point data identifies, to the computerized entity, the identity of the character. Upon determining that an embedded font document, such as a PDF document, does not include a set of code point data for a particular character, the character identification data is consulted to determine the identity of the particular character. In this way, a machine can recognize characters in the embedded font document and perform functions such as indexing or searching on the embedded font document.