Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1998-02-13
2001-01-23
Breene, John (Department: 2777)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000, C707S793000
Reexamination Certificate
active
06178434
ABSTRACT:
FIELD OF THE INVENTION
The current invention is generally related to a method and a system for generating a link for hypertext, and more particularly related to a method and a system for automatically generating a link between a legend for a figure and a corresponding main text and or a link between a figure and a corresponding text and or a legend.
BACKGROUND OF THE INVENTION
In the area of text image processing, there have been attempts for structuring components of textual information. One example of such a process is that a certain portion of text is made associated with another part of the text so that a reader can select relevant information based upon the textual association. Such text with internal associations is called hypertext. To accomplish such associations or links, certain languages such Hyper Text Markup Language (HTML) have been implemented.
Based upon the above described hypertext concept, one prior art attempt such as Japanese Patent Publication Hei 7-98708 discloses a system which selects relevant information from original text according to a reader's characteristics. For example, if the reader holds a certain position in a company, a predetermined type of information is available to him or her. By the same token, depending upon age and sex of a reader, certain predetermined information is displayed.
Another prior art attempt discloses a method of generating a HTML-based data. To generate the HTML data, text is scanned in as textual image, and based upon certain predetermined characteristics of the textual image, a certain portion of the text is associated with another portion of the text. “A Method of Automatically Generating HTML Based upon Image,” Fujii et al. Technical Report, Academy of Electronic and Communication, OSF 95-23, IE 95-55 (1995).
Lastly, a third and fourth prior attempts disclose similar methods of generating a HTML-based link between a figure and a corresponding portion of text. To establish a link, the text image is first tentatively divided into text segments and figure segments based upon relative positions within each column of the text image. Within a figure segment, a caption area is determined, and its characters are recognized. A certain predetermined set of characters such as “FIG.
1
” is detected in the recognized character data and is used as a label for the figure. Then, the label is further searched in the text segments, and the corresponding text portion is linked to the figure. “Generating Hyper Text From Printed Text for Electronic Library, ” Image Recognition and Understanding Symposium (MIRU), Ohira et al., (1996); “A Highly Applicable Method of Structuring Printed Text,” Koyama et al., Proceeding at Academy of Electronics Information and Communication, (1995).
The above described prior attempts generally fail to disclose a flexible automatic link generation method for text containing figures such as diagrams, tables, equations, drawings and etc. Although the above described third and fourth prior art attempts disclose a link generation between a figure and a corresponding portion of text, the link generation method requires a corresponding legend or caption area and relies upon predetermined labels or a set of characters. These requirements are not necessarily met for every text image.
SUMMARY OF THE INVENTION
In order to solve the above and other problems, according to a first aspect of the current invention, a method of generating a link for indicating a relationship in an image, includes the steps of: a) inputting the image containing segments, each of the segments containing elements; b) analyzing the image so as to generate layout analysis data for representing the elements in each of the segments; c) selecting an anchor which specifies at least one of the elements that is common among the segments; and d) generating a link between at least two of the segments based upon the anchor and the layout analysis data.
According to a second aspect of the current invention, a method of generating a link in an image, includes the steps of: a) inputting the image containing segments; b) dividing the image into segments, the segments containing a figure portion and a text portion, the figure portion and the text portion further containing a reference area; c) determining the reference area among the segments; d) recognizing characters in the image including the reference area; and e) generating a link between the reference area and a corresponding one of the text portion based upon the characters.
According to a third aspect of the current invention, a system for generating a link for indicating a relationship in an image, includes: an image inputting unit for inputting the image containing segments, each of the segments including elements; a layout analysis unit connected to the image inputting unit for analyzing the image so as to generate layout analysis data for representing the elements in each of the segments; an anchor processing unit connected to the layout analysis unit for selecting an anchor which specifies at least one of the elements that is common among the segments; and a link generation unit connected to the anchor processing unit for generating a link between at least two of the segments based upon the anchor and the layout analysis data.
According to a fourth aspect of the current invention, a system for generating a link in an image between a figure portion and a corresponding text portion, includes: an inputting unit for inputting the image; a segmentation unit connected to the inputting unit for dividing the image into segments, the segments containing a figure portion and a text portion, the figure portion and the text portion further containing a reference area; a reference area determination unit connected to the segmentation unit for determining the reference area among the segments; a character recognition unit connected to the inputting unit for recognizing characters in the image including the reference area; and a link generation unit connected to the reference area determination unit and the character recognition unit for generating a link between the reference area and a corresponding one of the text portion based upon the characters.
REFERENCES:
patent: 4723211 (1988-02-01), Barker et al.
patent: 5555362 (1996-09-01), Yamashita et al.
patent: 5659639 (1997-08-01), Mahoney et al.
patent: 5689585 (1997-11-01), Bloomberg et al.
patent: 5745360 (1998-04-01), Leone et al.
patent: 5761683 (1998-06-01), Logan et al.
patent: 5809317 (1998-09-01), Kogan et al.
patent: 5815830 (1998-09-01), Anthony
patent: 5848184 (1998-12-01), Taylor
patent: 5893127 (1999-04-01), Tyan et al.
patent: 7-98708 (1995-04-01), None
An Investigation of Automatic HTML Coding Method for Images, Fujii et al., Technical Report of IEICE OFS 95-23, IE 95-55 (1995) (Abstract).
Generating Hyper Text From Printed Text for Electronic Library, Image recognition and understanding Symposiun (MIRU) Ohira et al. (1996).
A Method of Logical Structuring of Printing Documents with High Applicability, Koyama et al.. College of Engineering, University of Osaka Prefecture (1995).
Breene John
Knoble & Yoshida LLC
Ricoh & Company, Ltd.
Robinson Greta L.
LandOfFree
Anchor based automatic link generator for text image... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Anchor based automatic link generator for text image..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Anchor based automatic link generator for text image... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2498688