2010-01-09

Google's Book Scanning Technology

clipped from news.cnet.com

Patent reveals Google's book-scanning advantage

Google has come up with a system that uses two cameras and infrared light to automatically correct for the curvature of pages in a book. By constructing a 3D model of each page and then "de-warping" it afterward, Google can present flat-looking pages online without having to slice books up or mash them onto a flatbed scanner.


clipped from patft.uspto.gov
United States Patent 7,508,978
Lefevere , et al. March 24, 2009

Inventors: Lefevere; Francois-Marie (Mountain View, CA), Saric; Marin (Palo Alto, CA)
Assignee: Google Inc. (Mountain View, CA)

Detection of grooves in scanned images

The diagram of Google’s patented technology is shown first and the diagram of the Japanese researchers’ technology is shown second. The similarity is immediately obvious.

Google's Book Scanning Patent

Diagram from article by researchers at U. Tokyo


blog it

clipped from news.cnet.com


Here's how the Google system is described in Patent 7,508,978:


First, the book is placed on a flat surface. Above it, an infrared projector displays a special mazelike pattern onto the pages.


Next, two infrared cameras photograph the infrared pattern from different perspectives.

This pattern can be shown on the book with infrared light; infrared cameras photograph it to deduce the 3D shape of the pages.


"The images can be stereoscopically combined, using known stereoscopic techniques, to obtain a three-dimensional mapping of the pattern," according to the patent. "The pattern falls on the surface of (the) book, causing the three-dimensional mapping of the pattern to correspond to the three-dimensional surface of the page of the book."

clipped from scitedaily.com

The University of Tokyo system works on the same principle, but only a single high-speed camera is used to serve the duties of all three cameras in Google’s system. It captures both pages simultaneously and also alternates between taking pictures under normal light, to capture the content of the page, and IR patterned light, to deduce the curvature of the page.


blog it

clipped from scitedaily.com
Google’s Book Scanning Music Patent

A related patent that Google was awarded on November 17 shows how they intend to use music to cue the human operator of their book scanning system. Flipping pages is tedious, and it is easy for people to slip up, perhaps skipping a page or accidentally taking pictures of their hand. The patent describes how a musical tone can be played from the speakers at regular intervals to give the operator a pace to flip pages to. It also describes how the system might be used play an error tone if the computer detects that a page may have been skipped (e.g. by detecting page numbers) or that the user’s hand may be in the picture.

Flipping pages is apparently quite hard. According to this site, sometimes people’s hands turn up in images on Google Books. Some companies offer robotic page flippers that do the task automatically.
clipped from www.youtube.com

4DigitalBooks

clipped from www.youtube.com

Robotic Book Scanning


blog it

Sources:
  1. Patent reveals Google's book-scanning advantage | Cutting Edge - CNET News
  2. United States Patent: 7508978
  3. Google’s Book Scanning Technology Revealed « SciTeDaily
  4. YouTube - 4DigitalBooks
  5. YouTube - Robotic Book Scanning
Related:
  1. The Secret Of Google's Book Scanning Machine Revealed - As A Matter Of Fact Blog : NPR
  2. Book scanning - Wikipedia, the free encyclopedia
  3. Optical character recognition - Wikipedia, the free encyclopedia
  4. Scan This Book! - New York Times