Working around OCR scanner

Skybird · 10-31-13, 11:36 AM

I want to scan some text from a very thick book, via OCR. But the book is so thick that the pages cannot be pressed onto the scanner's glass completely, the printed text also reaches too close to the inside limit of the page. The optics cannot recognise the text and cannot focus on it there.

So I wonder whether there is a way to have OCR being done not from live scanner footage, but doing a good HiRes photography via digital cam, and have the text extracted from that photo via OCR.

I only know FreeOCR, and have a ten year old copy of Omnipage somewhere. Don't want to spend money on this task.

OCR text extraction from an digital photo, bypassing flatbed scanner - any ideas, suggestions?

the_tyrant · 10-31-13, 01:21 PM

I know a guy who deals in antique books who swears by one of these:

http://www.fujitsu.com/us/services/c...nap/s1100.html

As for scanning with a camera, I know my phone has an app that can do this, maybe you can try looking for one for your phone?

Skybird · 10-31-13, 01:57 PM

I tried two Android apps for my tablet. Terrible results.

I need no hardware, I have a scanner, a very good one, even specialising for scanning photographic negatives and blowing them up. I need a software that extracts text from existing hi-res photos/pictures and turns it into editable text.

I mean if you feed text from a leaf of paper in a scanner, what happens is that the scanner creates an image first anyway - from which the OCR software then extracts the text.

Skybird · 10-31-13, 02:00 PM

Dang, maybe I had a slow day. I will try to photograph the book pages, print the images, and put these into the scanner then.

Hope this works.

10-31-13, 02:03 PM

There's alot of software that can convert digital images the way you want.

There's even some free on-line converters that do a nice job at times.

Platapus · 11-01-13, 08:29 AM

Our corporate library has a special (and very expensive) camera specifically designed for copying books. The software will automatically correct for the distortions caused by the bindings.

I have used it many times and I was surprised at how well it works. Well it should considering the cost.

10-31-13, 11:36 AM	#1
Skybird Soaring Join Date: Sep 2001 Location: the mental asylum named Germany Posts: 42,620 Downloads: 10 Uploads: 0	Working around OCR scanner I want to scan some text from a very thick book, via OCR. But the book is so thick that the pages cannot be pressed onto the scanner's glass completely, the printed text also reaches too close to the inside limit of the page. The optics cannot recognise the text and cannot focus on it there. So I wonder whether there is a way to have OCR being done not from live scanner footage, but doing a good HiRes photography via digital cam, and have the text extracted from that photo via OCR. I only know FreeOCR, and have a ten year old copy of Omnipage somewhere. Don't want to spend money on this task. OCR text extraction from an digital photo, bypassing flatbed scanner - any ideas, suggestions? __________________ If you feel nuts, consult an expert.

10-31-13, 01:21 PM	#2
the_tyrant Admiral Join Date: Jun 2010 Location: Canada Posts: 2,272 Downloads: 58 Uploads: 0	I know a guy who deals in antique books who swears by one of these: http://www.fujitsu.com/us/services/c...nap/s1100.html As for scanning with a camera, I know my phone has an app that can do this, maybe you can try looking for one for your phone? __________________ My own open source project on Sourceforge OTP.net KGB grade encryption for the rest of us

10-31-13, 01:57 PM	#3
Skybird Soaring Join Date: Sep 2001 Location: the mental asylum named Germany Posts: 42,620 Downloads: 10 Uploads: 0	I tried two Android apps for my tablet. Terrible results. I need no hardware, I have a scanner, a very good one, even specialising for scanning photographic negatives and blowing them up. I need a software that extracts text from existing hi-res photos/pictures and turns it into editable text. I mean if you feed text from a leaf of paper in a scanner, what happens is that the scanner creates an image first anyway - from which the OCR software then extracts the text. __________________ If you feel nuts, consult an expert.

10-31-13, 02:00 PM	#4
Skybird Soaring Join Date: Sep 2001 Location: the mental asylum named Germany Posts: 42,620 Downloads: 10 Uploads: 0	Dang, maybe I had a slow day. I will try to photograph the book pages, print the images, and put these into the scanner then. Hope this works. __________________ If you feel nuts, consult an expert.

11-01-13, 08:29 AM	#6
Platapus Fleet Admiral Join Date: Oct 2006 Posts: 19,362 Downloads: 63 Uploads: 0	Our corporate library has a special (and very expensive) camera specifically designed for copying books. The software will automatically correct for the distortions caused by the bindings. I have used it many times and I was surprised at how well it works. Well it should considering the cost. __________________ abusus non tollit usum - A right should NOT be withheld from people on the basis that some tend to abuse that right.

10-31-13, 02:03 PM	#5
Madox58 Stowaway Posts: n/a Downloads: Uploads:	There's alot of software that can convert digital images the way you want. There's even some free on-line converters that do a nice job at times.