Patent Shows Google Book Scanning a Musical Process

Sharing is caring!

Google was granted a patent today on one aspect of a book scanning process that raises the question what kind of music helps someone scan books best.

The patent is Pacing and error monitoring of manual page turning operator (US Patent 7,619,784), which lists Joseph K. O’Sullivan, R. Alexander Proudfoot, and Christopher R. Uhlik as inventors. Note the cameras and speakers above a book in a scanning cradle, in the image below from the patent:

Images from US Patent 7,619,784, granted to Google, showing a book in a cradle with cameras and speakers in a top image, and a person sitting at a manual scanning desk in a bottom image.

The patent was originally filed on June 30, 2003, and describes how a musical tempo might be used to help someone manually turning pages while scanning books. The abstract from the patent reads as follows:

Systems and methods for pacing and error monitoring of a manual page turning operator of a system for capturing images of a bound document are disclosed. The system includes a speaker for playing music having a tempo and a controller for controlling the tempo based on an imaging rate and/or an error rate. The operator is influenced by the music tempo to capture images at a given rate.

Alternative or in addition to audio, error detection may be implemented using OCR to determine page numbers to track page sequence and/or a sensor to detect errors such as object intrusion in the image frame and insufficient light.

The operator may be alerted of an error with audio signals and signaled to turn back a certain number of pages to be recaptured. When music is played, the tempo can be adjusted in response to the error rate to reduce operator errors and increase overall throughput of the image capturing system. The tempo may be limited to a maximum tempo based on the maximum image capture rate.

While we don’t know for certain that Google is using this process as described in the patent, there have been a few sites that have shown errors in Google Books that include images of hands turning pages.

The patent tells us of least three related patents from Google involving book scanning:

  • Moveable Document Cradle for Facilitating Imaging of Bound Documents
  • Acquiring and Using Three-Dimensional Information in a Document Scanning System
  • Imaging Opposing Bound Pages at High Speed Using Multiple Cameras

Why would Google manually scan books and magazines and other bound documents instead of using an automated process?

One reason would be to protect the documents themselves from flat-bed scanners that might damage the spines and bindings of books. The inventors behind the patent also tell us that automated page turning systems often require someone to manually turn the pages of books and magazines.

The patent goes into a fair amount of detail on how musical tempos might be used to help a manual scanner keep up a certain pace in turning pages, and how audio messages might be used to notify people scanning of errors on previous pages, including the appearance of hands and arms in images of those pages.

Of course, I’m curious as to whether people are listening to country, classical, hip hop, or rock, but we aren’t told in the patent if actual songs are used in the scanning of books.

Sharing is caring!

20 thoughts on “Patent Shows Google Book Scanning a Musical Process”

  1. That’s a pretty crazy thing to patent. Can you imagine somebody licensing this technology from them?

  2. I can’t believe there is a patent with such scanning technology but the only method they could think of to turn the pages is to use a manual operator turning pages to the beat of music.

  3. Hi Buzzlord,

    I’ve seen some scanning center operations, and they can be pretty labor intensive. While this appears to be unusual, I like the approach they are taking, especially considering some of the books that they are scanning are very rare. I also like the idea of audio cues that tell people that there have been some errors on previous pages. I don’t know if they would be open to licensing the technology, and I’m not sure whether someone would.

  4. Hi Google Terminator,

    We don’t know if all the books that Google is scanning are scanned manually, but it makes sense for them to have people scan the pages of books that might be fragile or rare (or both). I’m not going to say anything negative about the notion of scanning pages to music until I’ve been placed in that position myself. 🙂

  5. Pingback: » On the Google Chrome OS and other search engine news (Nov 22)
  6. I would be interested in what tracks they use for the operators, this really sounds interesting but also in a way sounds weird don’t you think?

  7. Hi seoyourblog,

    When I worked with Delaware’s Courts a few years ago, we had started a scanning program to begin moving our case records to electronic documents. We visited a couple of other agencies that had begun the transformation of records from paper to digital a few years before us, and it was surprising how labor intensive the whole process was, especially since many of our documents were odd sizes.

    When I see something like Google’s book scanning project, I’m staggered by the sheer magnitude of their undertaking.

    We hadn’t considered things like near-realtime error correction and audio warnings, or musical tempos to help pace manual scanning, but regardless of how weird it might sound, I think it does make some sense to consider things like that.

  8. They requested the patent in 2003?! Was the Google book scanning project even in the works back then? Just another example of the great forward thinking Google has going.

  9. Hi Frank,

    According to Google’s History of Google Books page, the Books project was officially launched in 2002:

    [2002]

    A small group of Googlers officially launches the secret “books” project. They begin talking to experts about the challenges ahead, starting with a simple but crucial question: how long would it take to digitally scan every book in the world? It turns out, oddly enough, that no one knows. In typical Google fashion, Larry Page decides to experiment on his own. In the office one day, he and Marissa Mayer, one of our first product managers, use a metronome to keep rhythm as they methodically turn the pages of a 300-page volume. It takes a full 40 minutes to reach the end.

    We’re also told on that page that in 2003, the company started “testing non-destructive scanning techniques,” and came up with some approaches that was “much gentler than current common high-speed processes.”

    Don’t know if that process was the one described in this patent, but the timing is right.

Comments are closed.