Google on Archiving and Retrieving Documents Using Your Camera Phone

You’ve returned to your hotel room from a business meeting with a pocket full of business cards from people that you’ve met, and receipts from your business trip. One at a time, you place the cards and receipts on a desk in your room and snap pictures of them with the phone on your camera, and send the photos off to Google to be processed.

The cards and receipts are scanned, and organized for you in your documents archive, so that they can be searched for, and used in your contacts list, and in your expense report.

That example brushes the surface of possibilities of a document archiving, storage, and retrieval system of images of physical documents, described in a new patent application published by Google.

Other documents that could be used in this type of system might include doctors prescriptions, tickets, contracts, and more. Depending upon how the system is set up, just taking a picture might trigger the document archiving system.

The basic underlying premise of the system is that “the myriad documents encountered in everyday life can be readily digitized, organized, stored and retrieved quickly and efficiently.”

While this seems to veer away from a way of organizing information on the web, it’s not too distant from the company’s stated mission: “Google’s mission is to organize the world’s information and make it universally accessible and useful.”

This system is described as being intended to be very easy to use – snapping a quick picture with your phone, which then can later be easily retrieved through a keyword search.

Optical character recognition (OCR) software, plus some processing techniques described in Google’s patent filing, can have the information captured in images sent to a personal computer to be archived and stored.

The patent application describes some of those processing methods in more detail, including ways that common problems associated with optical character recognition processing can be corrected using a language model.

The patent filing is:

Digital Image Archiving and Retrieval in a Mobile Device System
Invented by Krishnendu Chaudhury, Ashutosh Garg, Prasenjit Phukan, and Arvind Saraf
Assigned to Google
US Patent Application 20080126415
Published May 29, 2008
Filed November 29, 2006


A computer-implemented method of managing information is disclosed. The method can include receiving a message from a mobile device configured to connect to a mobile device network (the message including a digital image taken by the mobile device and including information corresponding to words), determining the words from the digital image information using optical character recognition, indexing the digital image based on the words, and storing the digital image for later retrieval of the digital image based on one or more received search terms.

9 thoughts on “Google on Archiving and Retrieving Documents Using Your Camera Phone”

  1. The latest version of Evernote does 99% of that. I take a photo of my receipt and/or business card on my phone, then e-mail it to a special Evernote address. It gets a very nice OCR treatment and every bit of it is searchable in the Evernote interface.

    It’s not as clean of an interface as I’d like and it doesn’t tie directly to contact lists or expense reports, but it’s quite close to what this article suggests is possible.

  2. Hi Mickey,

    Thanks for pointing out Evernote, and what they do.

    There definitely are some similarities, and some differences.

    As with most patent applications, what we are told is one example or illustration of what could be implemented, and likely doesn’t include all the details.

    The ideas are interesting ones, and it isn’t surprising that Google isn’t the only one looking to do this.

    What is interesting about Google’s patent filing is that they do provide a lot of details on the processing of images, so that the words on those images can be captured and indexed, so that they are searchable. I did take a look to see if evernote had a published patent application on a similar process, and though they had a number of patent filings, there weren’t any published that did describe something similar.

    Regardless, what is interesting here is that we get some insight from the patent filing on how a process like this might work. 🙂

  3. I think all the focus on searchable digital images is interesting. No doubt Google has big stakes and plans in this area. Whether it be recognizing text in digital pictures or people, shapes etc, image search is going to be the next big thing for search engines. After all – “a picture is worth a thousand words”.

  4. Hi peoplefinder.

    Have you been looking over my shoulder, at a couple of my next blog posts. 🙂

    I agree completely – the impact of image search should be pretty large in the next few years…

  5. As the others said this is a pretty nice idea and concentrating on image searches will be the future as People Finder said.

    You´ve a nice blog. I´ll take some time tomorow reading all of your posts. Keep it up.

    Best Regards

  6. @ Bill –

    I promise I have not been looking over your shoulder to get some inside info on your next blog posts, but I sure look forward to reading them.



Comments are closed.