OCR on a reader?

[recycledelectron]

It seems to me that one feature missing from most ebook readers is the ability to snap a pic of a page (think of a cell phone camera) and then OCR the image into a usable ebook file.

Let's say I am the note taker at a meeting, and need to pass around a copy of a proposal someone brought in on paper. I could snap a pic, OCR it, proof read it, and send it to every other ebook reader by email, bluetooth, wifi, or usb.

I think the iPhone / iPad can do something similar to this. Can any other devices do this?

Andy

[susan_cassidy]

Why would you bring something like that in only paper form? Email it to people. Also, I think what you would really want is the ability to email the text from the file, direct from the device, not go through a screen capture, then converting back to text. That's just backwards (if not crazy).

[CallOfCth'reader]

I would imagine that many multi-function devices, with various software and a camera, would be able to do it. I don't think that any dedicated e-readers will ever have those functions, because it would mean moving into the realm of multi-function devices (and raising prices for the extra hardware and software).

Personally, I wouldn't want an e-reader I bought to have a camera integrated, because that would mean I paying a higher premium for a function that I would never use: I already have multiple devices that can capture images, and a few that can OCR if I wish.

[Jellby]

That is a feature largely missing from desktop and laptop computers too (at least for my definition of "usable ebook file")

[Steven Lyle Jordan]

Originally Posted by susan_cassidy

Why would you bring something like that in only paper form? Email it to people. Also, I think what you would really want is the ability to email the text from the file, direct from the device, not go through a screen capture, then converting back to text. That's just backwards (if not crazy).

Welcome to the professional business world... where people still have their secretaries print out emails, so they can xerox them and hand them out at meetings. (Yes, I've met many of these. )

I think it would be a great idea to have any device handy that could quickly scan and OCR printed text. If it were mounted opposite the screen, you could possibly watch (and provide live corrections) as the camera and software did the OCR.

Better be careful about who you embarrass by converting their stuff to digital in front of them...

[Penforhire]

We're using our iPhones at work to capture white board images and there are multiple apps for converting them into PDF's and e-mailing them. Fairly awesome. I bought a house a few months ago and didn't have to use a FAX at all. I just did this sort of photo-scan to send my signed docs.

There are OCR software solutions for the iPhone but I haven't tried any. I mostly want handwriting recognition and that isn't ready for prime time. You might want to check out Google's free on-line PDF OCR service. Upload a PDF to your Google account and they'll use their servers to OCR it.

[DMcCunney]

Originally Posted by recycledelectron

It seems to me that one feature missing from most ebook readers is the ability to snap a pic of a page (think of a cell phone camera) and then OCR the image into a usable ebook file.

Let's say I am the note taker at a meeting, and need to pass around a copy of a proposal someone brought in on paper. I could snap a pic, OCR it, proof read it, and send it to every other ebook reader by email, bluetooth, wifi, or usb.

I think the iPhone / iPad can do something similar to this. Can any other devices do this?

Not that I'm aware of, but why would you want them to? The use case you give would be an at best infrequent occurrence.

Doing it would require a built in camera, which would add to device cost and not be wanted/needed by a lot of users, plus built-in OCR software and sufficient processing power to run it effectively, which some devices probably wouldn't have.

On the rare occasions I've needed to do something like what you suggest, I've scanned a hardcopy to my desktop, run OCR on it, converted to the desired format, and emailed those interested with a copy of the conversion after the fact. There was no reason to need to do so at the time the hardcopy was distributed. It could happily wait till a later convenient moment.

With any device, there is an implicit notion of what it will be used for, with the corresponding idea that uses beyond that are something else's job, and if you need to do one of them, you use something else. This is a "something else's job" case.
______
Dennis

[HamsterRage]

That reminds me of something I once saw happen in our office. An admin assistant printed a report, took the printed output to the scanner and scanned it, threw out the paper and converted the scanned input into a PDF.

At the time we had the technology (and we were already doing this all around the company) to simply produce the report in PDF and email it to the user, skipping the paper altogether. It just never occured to her, or anyone else involved in the process, to ask us to change the report over to PDF.

[Freeshadow]

methinks what he wants is kind of a combo of an ereader and a Cpen

(I have the lowest one (800C) on the old models list. they OCR .txt on-the-fly

[tomereader]

Sounds like ectaco k12

http://www.the-digital-reader.com/20...o-jetook-k-12/ http://www.coolcomputing.com/article.php?sid=4418

[DMcCunney]

Originally Posted by HamsterRage

That reminds me of something I once saw happen in our office. An admin assistant printed a report, took the printed output to the scanner and scanned it, threw out the paper and converted the scanned input into a PDF.

At the time we had the technology (and we were already doing this all around the company) to simply produce the report in PDF and email it to the user, skipping the paper altogether. It just never occured to her, or anyone else involved in the process, to ask us to change the report over to PDF.

When was this?

There are both a free add-on from MS and several third party efforts that can generate PDFs of Office documents. I wondered why MS didn't simply include the add-on with Office instead of requiring the user to go download it from them after the fact. The answer is apparently that if they included it with Office, they'd have to pay Adobe royalties. As a separately available download, they don't...
______
Dennis

[Penforhire]

It gets odder than that. Using Microsoft's add-in PDF output capability we can output PDF's from Word or Excel (MS lists 8 programs in their 'knowledge base') but not directly from Project! I had to install a 3rd party print-to-PDF driver (I like free PDF995).

[HamsterRage]

Originally Posted by DMcCunney

When was this?

There are both a free add-on from MS and several third party efforts that can generate PDFs of Office documents. I wondered why MS didn't simply include the add-on with Office instead of requiring the user to go download it from them after the fact. The answer is apparently that if they included it with Office, they'd have to pay Adobe royalties. As a separately available download, they don't...
______
Dennis

Well, this was a report coming from a back-end system, not from any MS Office or desktop application.

So it took work for us to figure out how to create PDF's on our Unix server where the legacy application was running and then send them to the user. Not that it was difficult, just something that had to be retrofitted. We use a nifty piece of software called PCL2PDF that does a very nice job of taking printer-ready output using PCL5 and turning that into PDF files.

But that's not really the point of the story. The admin never questioned the sanity of taking the paper, still warm from the printer, running it through the scanner and then shredding it! Even though she had to know that we were sending PDF reports all around the office (probably even to her).

[DMcCunney]

Originally Posted by HamsterRage

Well, this was a report coming from a back-end system, not from any MS Office or desktop application.

So it took work for us to figure out how to create PDF's on our Unix server where the legacy application was running and then send them to the user. Not that it was difficult, just something that had to be retrofitted. We use a nifty piece of software called PCL2PDF that does a very nice job of taking printer-ready output using PCL5 and turning that into PDF files.

Okay, makes sense. I've looked at PCL2PDF, but didn't have an actual use for it. (I'm a *nix admin, among other things.)

But that's not really the point of the story. The admin never questioned the sanity of taking the paper, still warm from the printer, running it through the scanner and then shredding it! Even though she had to know that we were sending PDF reports all around the office (probably even to her).

I've seen lots of that, though it's really bemusing when the ability to generate the PDF herself exists.

I got my first taste of it in the late '70's, working at a bank. One of my hats was resident expert in the mainframe based financial modelling system the bank used to generate the monthly expense numbers for management. The reports did comparisons: actuals better/worse than budget, actuals better/worse than forecast. The budget and forecast numbers were plugged in by the financial analysts in the areas. The actual numbers came from the bank's G/L system, also on the mainframe. Each month, the G/L first post would come out, and everybody got a printout. The analysts would then spend hours manually transcribing numbers from the printout into the appropriate place in the modelling system.

I spent a few weeks wandering around saying "Why are people manually re-entering this stuff when both pieces of software run on the mainframe? Why don't you program an interface from the G/L to the modelling system to do that automatically? Why are you paying MBA's $30K a year to be data entry clerks?"

Put that way, the light bulb came on over various heads in my region, and I was authorized to talk to DP about getting it done. The VP/Applications Development was delighted: it was a project that was doable and worth doing, and something his folks could get out the door and see meaningful results. They had been largely tasked to support Marketing, which could never figure out what it really needed, and whose requests tended to be ill-defined moving targets.

About half of what I did was serve as interface, translating user needs into things DP could understand. They had different mindsets and languages, so no surprise the obvious was overlooked.
______
Dennis

[elizilla]

My sig other has a funny story he tells, of when he was a computer operator in the days of timesharing. They had a report they ran every night. It was huge - two full paper boxes, which is what, ten thousand pages? A truck picked it up the same time every morning, around dawn, and ferried it to the customer site, so they've have it when they started work.

One night the printer was down, and they couldn't get it done before the truck left. So he was tasked with calling the user, first thing in the morning. Everyone thought the user would be upset. But here's what they said: "At the very end of the report, there's a two page summary. Can you please just fax it to me? That's all we ever look at."

D'oh!