Image-based retrieval for high quality visual or acoustic rendering

Patent metadata

Field	Value
Publication number	US-9244947-B2
Application number	US-201213535843-A
Country	US
Kind code	B2
Filing date	Jun 28, 2012
Priority date	Dec 29, 2006
Publication date	Jan 26, 2016
Grant date	Jan 26, 2016

Abstract

The embodiments of the invention relate to method for image-based retrieval of an image or text, comprising capturing a digital image; matching the digital image against digitally-stored documents; and retrieving one or more of the digitally-stored documents that match the digital image. Other embodiments relate to a device for image-based retrieval and rendering, comprising an image capturing device for capturing digital image; a device for processing the digital image to match it against digitally-stored documents; a device for retrieving a digitally-stored document that matches the digital image; and a device for rendering the selected retrieved image.

First claim

What is claimed is:

1. A method for image-based retrieval of an image or text, the method comprising:
extracting a plurality of distinctive local descriptors from the content of a digital image;
indexing a set of digitally-stored documents by extracting a plurality of distinctive local descriptors from the content of each document in the set;
matching the digital image against the set of digitally-stored documents by using near-duplicate image detection to compare corresponding distinctive local descriptors of the digital image and each document in the indexed set, wherein near-duplicate image detection comprises using locality-sensitive hashing;
retrieving a digitally-stored document from the set of digitally-stored documents that matches the digital image, the retrieved digitally-stored document comprising a composite of images of a plurality of articles in a defined layout, wherein each of the plurality of articles has a corresponding article layout;
after said matching and said retrieving the digitally-stored document, retrieving, from the composite of images, one article from the plurality of articles that corresponds to the digital image based on overlapping areas associated with the plurality of articles, wherein for each article of the plurality of articles, the overlapping area is between the boundary of the corresponding article layout of the article and the boundary of the complete digital image, the retrieved article having the greatest overlapping area among the overlapping areas; and
rendering the retrieved article for output to a user device, wherein rendering includes at least one of: converting the retrieved article to text and converting the text to speech, and converting the retrieved article for Braille output.

2. The method of claim 1 , wherein rendering includes converting the retrieved article for visual display.

3. The method of claim 1 , wherein rendering includes converting the retrieved article for audio output.

4. The method of claim 1 , wherein converting the retrieved article for Braille output includes converting the retrieved article to text.

5. The method of claim 1 , further comprising capturing the digital image using a digital camera or a scanner.

6. The method of claim 1 , further comprising rendering the retrieved digitally-stored document.

7. A system for image-based retrieval and rendering, the system comprising:
a first physical device comprising hardware, the first device extracting a plurality of distinctive local descriptors from the content of a digital image;
a second physical device comprising hardware, the second device indexing a set of digitally-stored documents by extracting a plurality of distinctive local descriptors from the content of each document in the set;
a third physical device comprising hardware, the third device matching the digital image against the set of digitally-stored documents by using near-duplicate image detection to compare corresponding distinctive local descriptors of the digital image and each document in the indexed set, wherein near-duplicate image detection comprises using locality-sensitive hashing;
a fourth physical device comprising hardware, the fourth device retrieving a digitally-stored document from the set of digitally-stored documents that matches the digital image, the retrieved digitally stored document comprising a composite of images of a plurality of articles in a defined layout, wherein each of the plurality of articles has a corresponding article layout;
a fifth physical device comprising hardware, the fifth device retrieving, after the digital image is matched and the digitally-stored document is retrieved, from the composite of images, one article from the plurality of articles that corresponds to the digital image based on overlapping areas associated with the plurality of articles, wherein for each article of the plurality of articles, the overlapping area is between the boundary of the corresponding article layout of the article and the boundary of the complete digital image, the retrieved article having the greatest overlapping area among the overlapping areas; and
a sixth physical device comprising hardware, the sixth device rendering the retrieved article for output to a user device, wherein rendering includes at least one of: converting the retrieved article to text and converting the text to speech, and converting the retrieved article for Braille output.

8. The system of claim 7 , further comprising an image capturing device comprising hardware, the image capturing device capturing the digital image.

9. The system of claim 8 , wherein the image capturing device includes a digital camera or a scanner.

10. The system of claim 7 , further comprising the sixth device rendering the retrieved digitally-stored document.

11. The system of claim 7 , wherein the sixth device rendering the retrieved article for output to a user device further includes converting the retrieved article for visual display.

12. The system of claim 7 , wherein the sixth device rendering the retrieved article for output to a user device further includes converting the retrieved article for audio output.

13. The system of claim 7 , wherein converting the retrieved article for Braille output includes converting the retrieved article to text.

Assignees

• Thong J M Van
• Sukthankar Rahul
• Intel Corp

Inventors

• Thong J M Van
• Sukthankar Rahul

Classifications

• G06F17/30277 Primary

  Physics · mapped topic

• G06F17/30011

  Physics · mapped topic

• Y10S707/917

  Text · CPC title

• G06F16/93

  Document management systems · CPC title

• G06F16/532 Primary

  Query formulation, e.g. graphical querying · CPC title