Reformatting documents using document analysis information

• US 20080037873A1
• Filed: 08/16/2007
• Published: 02/14/2008
• Est. Priority Date: 01/29/2003
• Status: Active Grant

• Current Assignee: Christophe Marle, Edward Schwartz, Kathrin Berkner, Michael Gormish
• Original Assignee: Christophe Marle, Edward Schwartz, Kathrin Berkner, Michael Gormish
• Inventors: Berkner, Kathrin, Marle, Christophe, Gormish, Michael, Schwartz, Edward
• Granted Patent: US 7,792,362 B2
• Time in Patent Office: Days
• US Class Current: 382/176
• CPC Class Codes: G06F 16/9577 Optimising the visualization of content, e.g. distillation of HTML documents
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text

Abstract

A method and apparatus for reformatting electronic documents is disclosed. In one embodiment, the method comprises performing layout analysis on an electronic version of a document to locate text zones, assigning attributes for scale and importance to text zones in the electronic version of the document, and reformatting text in the electronic version of the document based on the attributes to create an image.

Claims

1. A method comprising:

generating a multiresolution segmentation image for an electronic version of a document;

performing a connected components analysis on the multiresolution segmentation image to generate a list of image connected components, along with their locations within the multiresolution segmentation image and multiresolution bit distribution;

performing layout analysis on the electronic version of a document to locate text zones;

assigning attributes to text zones in the electronic version of the document;

creating a list of text components associated with the text zones; and

merging component images associated with the image connected components of the multiresolution segmentation image and the text components.

2. The method defined in claim 1 further comprising creating a component image of text zones using bounding boxes of layout analysis.

3. The method defined in claim 1 further comprising obtaining attributes for each of the image components in the list of connected components.

4. The method defined in claim 3 wherein the attributes include the resolution of the component.

5. The method defined in claim 3 wherein the attributes include the location of the component.

6. The method defined in claim 3 wherein the attributes include an importance value for the component.

7. The method defined in claim 3 wherein the attributes include an indication of the smallest rectangle that contains the component.

8. The method defined in claim 3 wherein merging the component images comprises computing the overlap between each image component with the text components.

9. The method defined in claim 1 further comprising subtracting at least one text component from an image component if an amount of overlap is greater than a first threshold and less than a second threshold.

10. The method defined in claim 1 wherein merging is based on one or more attributes.

11. The method defined in claim 10 wherein the one or more attributes comprise an importance value.

12. The method defined in claim 10 wherein the one or more attributes comprise an importance from analysis of layout for text components.

13. An apparatus comprising:

means for generating a multiresolution segmentation image for an electronic version of a document;

means for performing a connected components analysis on the multiresolution segmentation image to generate a list of image connected components, along with their locations within the multiresolution segmentation image and multiresolution bit distribution;

means for performing layout analysis on the electronic version of a document to locate text zones;

means for assigning attributes to text zones in the electronic version of the document;

means for creating a list of text components associated with the text zones; and

means for merging component images associated with the image connected components of the multiresolution segmentation image and the text components.

14. The apparatus defined in claim 13 further comprising means for creating a component image of text boxes using bounding boxes of layout analysis.

15. The apparatus defined in claim 13 further comprising means for obtaining attributes for each of the image components in the list of connected components.

16. An article of manufacture comprising one or more recordable media containing executable instructions that, when executed by a system, cause the system to:

generate a multiresolution segmentation image for an electronic version of a document;

perform a connected components analysis on the multiresolution segmentation image to generate a list of image connected components, along with their locations within the multiresolution segmentation image and multiresolution bit distribution;

perform layout analysis on the electronic version of a document to locate text zones;

assign attributes to text zones in the electronic version of the document;

create a list of text components associated with the text zones; and

merge component images associated with the image connected components of the multiresolution segmentation image and the text components.

17. The article defined in claim 16 further comprising instructions which when executed by the system cause the system to create a component image of text boxes using bounding boxes of layout analysis.

18. The article defined in claim 16 further comprising obtain attributes for each of the image components in the list of connected components.