ScanBiz Mobile Solutions is a 3 person company, based in New York. We love the iPhone as a consumer device as well as its development environment, and with the release of the iPhone 3GS in June 2009, we felt that the increased processor speed and better camera finally delivered a platform able to do more - faster - with images, right on the device itself. Optical Character Recognition (OCR) is one of the tasks now enabled on the iPhone.
We are surrounded by text everywhere: window signs, commercial logos and phone numbers plastered on trucks, flyers, take-away menus - and yet to capture and use all this information we essentially ressort to typing these phone numbers and web sites manually into a phone or computing device. We thought we should help change that, with the help of the mobile phone camera and OCR applications extracting the relevant textual information in these images. In fact, Google thought so as well! You can read about their recent unveiling of Google Goggles on the Android mobile platform.
We use the Tesseract OCR engine (version 3.0). Tesseract is an open source optical character recognition engine. It was originally developed as proprietary software at Hewlett-Packard between 1985 until 1995. In 2005, Hewlett Packard and UNLV released it as open source in 2005. Tesseract is currently developed by Google and released under the Apache License, Version 2.0. The Tesseract OCR engine was in the top 3 in terms of character accuracy in 1995 and still does pretty well compared to commercial alternatives to this day. You can upload any image to see what text Tesseract recognized at www.free-ocr.com. Good character recognition depends heavily on getting clean images where the text is clearly differentiated from non-text elements, colors and shadows. This is why a good image processing algorithm is required prior to performing OCR: we use an excellent and innovative algorithm provided by a team of researchers at the CMATER Research Center, Jadavpur University, India
Our first product, ScanBizCards Business Card Reader for the iPhone 3GS, specifically targets these little cardboard rectangles choke full of information which we all swap between friends and colleagues, or collect at local service providers, bars and restaurants: business cards. Having said that, for folks like us obsessed with scanning business cards of all shapes and sizes, "everything looks like a business card" such as this street sign ... and ScanBizCards definitely lends itself to scanning such images and use the information in them just like a 3"x2" business card! If you notice someone taking a picture of the "1-800-AM-I-SAFE" sign at the back of your car, do not be alarmed, it may just be one of us roaming the streets and testing ScanBizCards on any phone number, email address or URL in sight.
We launched ScanBizCards V1.0 on October 28, 2009 and have been busily adding one feature after another (see complete list at www.scanbizcards.com/features) above and beyond just scanning a card and adding the information into the address book - scanning a card is just the beginning or a potential fruitful new rapport and we'd like to help networkers and business professionals make these connections happen faster and easily. Special thanks go to J Grant Design for putting a friendly look on our techie work.
We'll let you decide whether scanning an image of a business card in seconds then just touching a phone number to dial it qualifies as augmented reality, meanwhile it's pretty useful and is a sure way to impress your friends at a cocktail party!
Check out the ScanBizCards Blog discussing updates to the ScanBizCards iPhone application.
At the time of this writing, the competition for leadership in business card iPhone scanning applications is fierce, with usually 3 or 4 such applications within the top 20 iPhone apps in the business category, all of which doing an excellent job at scanning a card image quickly and letting you add the text recognized into a new contact in your address book. What truly sets ScanBizCards apart is that "scanning & adding to the address book" is the end of the road with all other apps - for us "scanning is only the beginning" and we offer a long list of extended features to managed and act on scanned cards. The gist of these features is:
- Manage: enable the user to retain and manage the card images as such, along with the recognized text: business cards serve as visual cues to remind us who the contact is and where we met, they contain much more information than just what we chose to include in the address book. ScanBizCards presents options to store cards in custom folders, sort them by name/company/date, flip though them in list view of "cover flow" 3D view and search all cards for text.
- Act: empower the user to act on a new card on the spot and without leaving the app: send a Quick Intro email with one's own contact info, forward the information to a colleague, set a calendar follow-up reminder, send a LinkedIn invite.
- Cloud Sync: an optional feature we call "Web Sync" instantly saves each new card to the cloud for safekeeping and lets users access their cards from anywhere via the browser. Web Sync also allows keeping multiple devices in sync in real-time.
- Export: scanned cards may be exported to an Excel CSV (to import for example into a CRM back-end like Salesforce.com, ACT or Sugar CRM) or uploaded to Evernote.
We would love to hear your thoughts on what other OCR applications you'd like to see on mobile devices, including possible joint developments - please contact us.