Methods and systems for automatic extraction and retrieval of auxiliary document content
AssigneeSharp Laboratories of America, Inc.
InventorAhmet Mufit Ferman
Life of this patent
Estimating the term…
Abstract
Aspects of the present invention are related to systems and methods for automatically extracting, from a document image, references to relevant external content and automatically retrieving the external content associated with the references.
Claims17
A method for retrieving auxiliary document content indicated in a document image associated with a United States Patent and Trademark Office patent document front page, said method comprising:
- performing layout analysis on a document image, thereby determining a first region-of-interest, wherein said performing layout analysis comprises:
- identifying a header region in said document image;
- identifying a main body region in said document image;
- and
- identifying a demarcation line between said header region and said main body region;
- extracting, from said first region-of-interest in said document image, a first reference to a first external content;
- and
- retrieving said first external content.
A method for retrieving auxiliary document content indicated in a document image associated with a United States and Trademark Office patent document front page, said method comprising:
- performing layout analysis on a document image, thereby determining a first region-of-interest, wherein said determining said first region-of-interest comprises locating Internationally agreed Numbers for the Identification of (bibliographic) Data code “56” in said document image;
- extracting, from said first region-of-interest in said document image, a first reference to a first external content;
- and
- retrieving said first external content.
A method for retrieving cited United States Patent and Trademark Office (USPTO) documents indicated in a USPTO patent document, said method comprising:
- locating a header region in an image of a first page of a USPTO patent document;
- locating a main body region in said image;
- locating an Internationally agreed Numbers for the Identification of (bibliographic) Data (INID) code “56” in said main body region;
- identifying a region-of-interest based on said located INID code “56”;
- extracting a first reference from said region-of-interest;
- generating a USPTO database query comprising said extracted first reference;
- and
- obtaining an auxiliary document using said generated USPTO database query.
Dependencies are inferred from the claim text.