Is Mobile Capture Only Hype?

Community Topic(s):

Keywords: OCR, capture, mobile, mobile applications, mobile device, data capture

Current Rating:
(1 ratings)

Last week, Chris Riley discussed a topic very near and dear to my heart – mobile capture. I respect and agree with Chris’ opinion, however, I wanted to take a closer look at his post, “Mobile Capture: The hype, the truth, the future.” In some cases I will agree, in others I will bring up a counterpoint. My goal in either scenario is to continue the debate and show you why you should care.

As most of you have read, I am passionate about the potential of mobile capture. I have also addressed the mobile capture issue numerous times for the AIIM Capture community. (If you’ve missed any of these previous posts, feel free to check out commentary on CTIA, predictions for mobile capture in 2011, how mobile is enabling the cloud, and lastly, mobile capture as the next kiler app.) In this post, I will be addressing specific points made by Chris.

In the hype section of Chris’ post, he says that the human element of mobile capture is a stumbling block. I agree that the user plays a large part in enabling a quality scan. Whether it is by seeking appropriate lighting or having a steady hand when using the smartphones camera, all of these factors impact the result. However, I don’t think this is something we can’t work around.  Hardware is only getting better with higher resolution cameras, auto-focus, built in flash, etc. and where it doesn’t improve, software must. This could take the form of better identification of questionable characters, or better image pre-processing. These are areas where we as solution providers can deliver additional value.

I know what you’re thinking: “all of these steps combined won’t make mobile capture as accurate as a dedicated scanning environment.” Well, you are probably right, but the real question is: Does it need to be? The business and consumer use cases for mobile capture are, and should remain, very different than those of a dedicated scanning environment. Mobile capture is good for processing small bits of information and communicating it faster. It is not designed for OCRing full documents. That is still the domain of dedicated scanning environments. The two technologies should instead be complimentary, and work together to extend information flow to the furthest reaches of the business.

Later in his post, Chris makes a very interesting statement: “… the typical user will be using OCR, but have no idea what OCR is.” This is interesting for a couple reasons. If you ask traditional OCR users today, I bet that most do not know what technology makes their paper documents digital.  This is the sign that OCR has improved over the past decade, and people now see it as something that “ just works.”  There are some OCR providers that spend more money on R&D than others, and  with technology continuing to improve, capture providers can spend more time on collaborating with application developers on on how the overall solution can be enhanced.  

As I included in my previous post, “The Capture Bowl,” many of the leading analyst firms see smartphone growth accelerating even faster than the 96 percent increase we realized in 2010. They also predict that data capture’s future will be driven by mobility. This means we will be forced to use our technology in new ways to deliver innovative solutions to smartphone users. And, whether the user knows what OCR is or not, we have the opportunity to change the way people process the information around them.

What are your thoughts on mobile data capture? Do you think that the technology can overcome the limitations introduced by untrained end users? Is it important that end users understand or even know they are using OCR? I would love to hear your thoughts in the comments.

Report

Rate Post

You need to log in to rate blog posts. Click here to login.

Add a Comment

You need to log in to post messages. Click here to login.

Comments

Daniel O'Leary

Use it ALL the time

I've become a primary user of mobile capture; scanning receipts, business cards, and customer data.If only Abbyy was integrated into more apps!
Report
Was this helpful? Yes No
Reply
Chris Riley, ECMp, IOAp

Joe can fix that

I'm sure Joe will like to fix that for you. Daniel about causing trouble at AIIM, we should arrange post smack downs! Bryant?
Report
Was this helpful? Yes No
Reply
Bryant Duhon

Smack Downs, eh?

Sounds similar to an idea I've had for a while to do your basic point/counterpoint discussion. No eye gouging, fish hooking, or blows below the belt; after that, maybe no rules :)
Report
Was this helpful? Yes No
Reply

Chris Riley, ECMp, IOAp

I called it

Joe,

I was hoping to elicit some response from you, and thank you for your post. There is no question the benefit and the future of mobile capture. What I was addressing was a bit more subtle. Vendors of technology very often miss the connection between perceived value of technology and the reality of us. My article was an attempt to bridge that gap for both the vendors and end-users. The subtle message was, there is danger in just getting excited about a technology without understanding it's true capabilities, or understanding where the true value is. I will use an example. BCR. Does snapping a business card provide true value to just an individual or is there potential for it adding benefit to the entire organization? When BCR is demoed I'm hard pressed to find someone after the demo that does not want it. But when I survey users of BCR I found that a typical user is not willing to correct more then one error in any capture. I also found that users who had to repeatedly correct errors ( one or more errors for group of business cards) soon dropped the technology in favor of more manual approaches to entering business card information. So the excitement drives adoption, but disillusionment ruins the reputation of a technology.

The reality is that if you are only correcting one error on a business card there is still a time savings so why would people drop the technology in favor for something slower? It's all becuase the perceived value did not live up to their expectations, and they know what to expect with manual entry. You can see in this scenario that this neither benefits the user, or vendor. User is walking away from something that does benefit them, vendor is losing a customer. All becuase of "hype".

The value of mobile capture is clear and dear to my heart as well. What I encourage users to do is get past what they "want, dream, hope" the technology does to what it really does with testing. Then quantify it's value. I don't attempt to discourage mobile capture usage, but to better align expectations which benefits everyone.

My survey has been on going since 2007 and totals now 67 current and past BCR users.
Report
Was this helpful? Yes No
Reply

Joe Budelli

Yes and yes

I fully agree that we should engage in a friendly battle with some basic point/counterpoint discussions about what we see at AIIM and post show. Chris, I am glad you knew how to entice me and get a response, as I thoroughly enjoyed your post and thought it brought valuable questions and ideas to the table. I think the main point (to both of our ends) is that we need to encourage the value of mobile capture technology. While aligning end users expectations and outcome is ideal, it is our job to continue to deliver new innovative solutions and change the way people process information. I think that through this continued innovation and improvements to the core technology, we will get to a point where the hype and actual performance are the same.
Report
Was this helpful? Yes No
Reply

Michael Spang

Data that you can use....

After years of talk mobile capture is becoming a legitimate technology. Yes, scanning business cards could be a "fun app" for awhile.... but Chase Bank seems to think that there is a market for check scanning with a smart phone. They have invested millons in advertising. The long term success or failure of mobile capture depends on whether or not technology will be developed to the point where it makes data more useful.
Report
Was this helpful? Yes No
Reply

Vasily Loginov

A "QA" for mobile scanning

I just want to add that modern imaging technologies already can solve quality issues for mobile scanning. In digital photo industry during last ten years were developed powerful automatic enhancement routines that successfully work in photo kiosks and labs. Fortunately, it is known how to adapt them to «DOCUMENT» scenes.

One example – adapted photo technology can automatically convert darken, noisy document to conventional quality scan without ANY characters degradation. The only requirement is that the routine should be tuned to particular camera chip.

Common approach, applied to mobile scanning, is to supply automatic processing with mobile scanning profiles for different shooting conditions, devices, document types and document usage. I can imagine profiles for “BC OCRing”, “ shaded document faxing”, “saving multicolor document printing»…

Automatic (visual, OCR, compression …) quality assessment may be also very valuable function as an adviser to miss a scan or shout another copy. Again, there is no technology barrier to include the function into current mobile scanning applications.

Mobile devices bring into data exchange new types of “documents” like short texts from objects in non-document environment. In these cases OCR technology may work with insufficient accuracy. Again, modern image processing has special image preparation techniques to assist in overcoming this. Mobile HD video capture and modern video technology allow restoring OCRable texts even from very poor, night shaky videos.
Report
Was this helpful? Yes No
Reply

This post and comment(s) reflect the personal perspectives of community members, and not necessarily those of their employers or of AIIM International