Data format conversion

• US 20070009161A1
• Filed: 06/30/2006
• Published: 01/11/2007
• Est. Priority Date: 07/08/2005
• Status: Active Grant

• Current Assignee: Cynsight, LLC
• Original Assignee: Cynsight, LLC
• Inventors: Hollingsworth, William
• Granted Patent: US 10,528,806 B2
• Time in Patent Office: Days
• US Class Current: 382/229
• CPC Class Codes: G06V 30/40 Document-oriented image-based pattern recognition

Abstract

Converting intermediate document data representing document text derived from data in an image data format into a semantically-meaningful tagged text data format may be provided. Intermediate document data derived from document image data may be imputed. The intermediate document data may comprise character data corresponding to characters in the document and attribute data corresponding to one or more attributes of characters in the document. The intermediate document data may then be processed according to attribute-dependent rules. Tagged text input data may be generated comprising tagged section of the document text. The tags may define semantically meaningful portions of the text determined according to the attribute data.

Claims

1. A method of converting intermediate document data representing document text derived from data in an image data format into a semantically-meaningful tagged text data format, the method comprising:

inputting intermediate document data derived from document image data, said intermediate document data comprising character data corresponding to characters in the document and attribute data corresponding to one or more attributes of characters in the document;

processing the intermediate document data according to attribute-dependent rules; and

generating tagged text data comprising tagged sections of said document text, the tags defining semantically meaningful portions of said text determined according to said attribute data.

2. The method of claim 1 wherein said attributes of characters in the document include one or more of font type, font size, line spacing, character bold, character italicised, character underlined, and character position.

3. The method of claim 1 wherein said processing comprises:

reading said intermediate document data; and

operating on data derived from said intermediate document data using rule data defining a plurality of text identification rules, said rules comprising rules for identifying a set of semantically meaningful portions of said document text, to generate said tagged text data.

4. The method of claim 3 wherein said rule data for a said rule comprises computer program code for identifying a set of characters for a said portion of document text from said character attributes and/or one or more variables defined by a said rule.

5. The method of claim 4 wherein said operating comprises causing said program code to be executed, said code including code to generate one or more commands dependent upon one or more variables dependent upon a said character attribute;

and then executing said commands.

6. The method of claim 5 wherein said commands include one or more commands to set values of one or more of said variables dependent upon a said character attribute.

7. The method of claim 3 wherein at least some of said rules define a hierarchy of semantically meaningful portions of said document text.

8. The method of claim 3 wherein said rules include one or more of a zone rule to identify a zone of said document text, a paragraph rule to identify a paragraph of said document text, a line rule to identify a line of said document text, a title rule to identify a title of said document text, an author rule to identify an author of said document text, a caption rule to identify a figure caption of said document text, and a footnote rule to identify a footnote of said document text.

9. The method of claim 3 further comprising inputting user data to define one or more of said rules of a template data file.

10. The method of claim 1 wherein said tagged text data comprises data in a markup language format.

11. The method of claim 1 further comprising converting said tagged text data to speech data for audio output.

12. The method of claim 11 further comprising selecting from among said semantically meaningful portions of text, portions to convert to said speech data.

13. The method of claim 1 further comprising inputting said document image data, and generating said intermediate document data from said document image data.

14. A method of converting a document from an image format into a semantically-meaningful format, the method comprising:

receiving document image data;

generating intermediate document data comprising words, lines, paragraphs and zones by generating character data corresponding to characters in the document and attribute data corresponding to one or more attributes of characters in the document and grouping said character data;

processing the intermediate document data according to format-identification rules and generating output data comprising tagged sections of text, the text corresponding to text in said document, the tags defining portions of said text determined by said attribute data; and

outputting said output data.

15. A data carrier carrying a template data structure, said template data structure comprising data for a plurality of data format conversion rules for application to intermediate document data derived from document image data, said intermediate document data comprising character data corresponding to characters in the text of the document and attribute data corresponding to one or more attributes of characters in the document, a said rule comprising data for identifying a set of characters for a portion of document text from said character attributes and/or one or more variables defined by a said rule.

16. The data carrier of claim 15 wherein said rule comprises computer program code for identifying a semantically meaningful portion of said document text.

17. The data carrier of claim 15 wherein at least some of said rules define a hierarchy of semantically meaningful portions of said document text.

18. The data carrier of claim 15 wherein said rules include one or more of a zone rule to identify a zone of said document text, a paragraph rule to identify a paragraph of said document text, a line rule to identify a line of said document text, a title rule to identify a title of said document text, an author rule to identify an author of said document text, a caption rule to identify a figure caption of said document text, and a footnote rule to identify a footnote of said document text.

19. The data carrier of claim 15 further comprising user interface computer program code to implement a user interface for user definition of at least one said rule.

20. The data carrier of claim 15 further comprising an interface to a data processing system to process said intermediate document data according to character attribute-dependent rules and to generate tagged text data comprising tagged section of said document text, the tags defining semantically meaningful portions of said text determined according to said attribute data.

21. A data carrier carrying a list of at least one attribute-dependent rule for use in processing document data, the document data comprising character data corresponding to characters in a document and attribute data corresponding to attributes of characters in the document, the rule comprising:

a first portion in an imperative programming language for determining portions of said document delineated by said attribute data; and

a second portion in a document processing language for generating output data comprising tagged sections of text, the tags defining portions of said text delineated by said attribute data.

22. An apparatus for converting a document from an image format into a semantically-meaningful format, the apparatus comprising:

a data memory operable to store data to be processed;

an instruction memory storing processor implementable instructions; and

a processor operable to read and process the data in accordance with instructions stored in the instruction memory;

wherein the instructions stored in the instruction memory comprise instructions for controlling the processor to;

receive document image data;

generate from said image data intermediate document data comprising character data corresponding to characters in the document and attribute data corresponding to one or more attributes of characters in the document;

process the intermediate document data according to attribute-dependent rules and generate output data comprising tagged sections of text, the text corresponding to text in said document, the tags defining semantically meaningful portions of said text determined by said attribute data; and

outputting said output data.

23. The apparatus of claim 22 further comprising a speech synthesiser configured to receive said output data and to convert said output data to audio speech data.