Patent reveals Google's book-scanning advantage
Google is cagey about exactly how it scans books for its digital library effort, but a patent reveals details--and the hurdles competitors face.
Sometimes overlooked in theis any consideration of the mechanics of economically scanning the books in the first place, but a patent awarded to Google gives insight into how the search behemoth accomplishes the task.
In short, 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.
The sophistication of the technology illustrates that would-be competitors who want to feature their own digitized libraries won't have a trivial time catching up to Google, which already has scanned more than 7 million books. Any unskilled laborer can plop a book on an ordinary scanner and run some optical character recognition (OCR) operations that convert the imagery into textual data, but doing so rapidly and with high-quality images is another matter.
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.
"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."
Next, photos of the page taken with conventional cameras can be de-warped, permitting easier OCR and a better image when showing the real book in conjunction with search results based on the text.