I don’t like the fact that PDFOCR_Output is created in the root C\Directory even the program itself was installed in C\program files and there is no way to change the output directory.
I dont like the output text files open in wordpad instead of pdf_ocr editor shown on the tutorial.

Thanks #23 Giovanni! Lots of low quality results with this program.

I wonder if many of the install/register problems are because of the GAOD wrapper to keep you from re-installing it elsewhere. The wrapper often bloats the program size tremendously. Who knows what all these limiters do to your registry? Some of mine and other people's programs have been "un-registered" Headache city. Seems odd since other programs from this company seem to work quite well.

Most OCR doesn't work well (except simple words and basic fonts). Add diacritical marks, other languages - now you're asking for a brilliant complex program and pushing it. Some company should make a great PDF - OCR to text converter, but it needs to be VERY smart and accurate. Speech to text is another area not there yet, despite Dragon's best efforts.

Cell phones with Droid apps seem to do a good job of speech to text. However emails you send from a cell phone are usually small words and a simple vocabulary, That's easy. Difficult technical documents with photos, columns, and formulas with sub or super scripts are hard to do correctly. Most HUMAN BEINGS don't know what the heck to do with Scientific NOTATION! LOL.

This has nothing to do with "OCR" - the text in every .PDF file is available as text right from the file menu of Acrobat reader (the free version) - there's nothing special about software that extracts text from a .PDF. Thumbs-down on this one .

Followed the suggestions by 37 for registration in root drive and finally registered. Will try to use tomorrow and uninstall if as poor as folks are saying.

I am happy with cute PDF Writer http://www.cutepdf.com/ been using it for years

Hi,

I have installed theall, I er inst prgram, however there is a problem. After it is installed I did the registration thing, and it does not seem to register. Every time I run the program it asks me to register it. Then it asks me to restart the program....this is a continuous loop. That is for whatever reason it apparently does not actually register even though it says it is registered. I have uninstalled and re-installed and that has not helped.

Can someone give me some idea on how to get rid of this problem?

Thanks for whatever help is offered.

Cheers

Dave
PS and of course the 24 hrs of GOTD is almost up so I am not sure if it will even work after that ends.

oh my!
a program that could be of immense use to many and especially to the blind out here, and the interface is totally inaccessible to us!

what a laugh,

to whiterabbit (# 12), go take a look at abbey fine reader, runs about 99 dollars here, but there are older versions available for less and it works very well on OCR'ing image documents. far less expensive though not as accessible as the OCR applications for the blind one runs 999.95, and the other is in the 12000 dollar category.

you can also try out a application that can be googled, called freeOCR

take care,
blind computer user

Not For Me (I use something better)

Many people seem to mistakenly think that this is for extracting text from a PDF.
It is for converting a photo/scanned page into a searchable/copyable document.

A better offer than the last few days, and potentially useful.
Based on reviews here today, I will not be downloading.
I use OmniPage Pro v17 (by Nuance Communications)

It's paid-for software, it's expensive, it's not always intuitive to use if you want to tweak things, but it's brilliant.

-you can extract PDF's to almost any format (RTF, Word, and LOTS of others)
-You can convert images to searchable PDF's
-It will automatically tell the difference between images and text.
-It will automatically rotate pages (it uses the text it's found to decide which way is "up")
-you can scale, crop, correct, etc.
-it will ask you about anything that it is unsure of
-It can save in continuous or true-page (ie website vs book)
-it can scan in & convert directly from any P+Play scanner (can be 'interesting' to set up though)
-you can drop a whole bunch of things in a folder and batch convert, either each to a separate file, or contiguously

I use it mostly with an android app "cam-scanner". I use my phone's camera to grab the pages and then save all those images as a PDF. I copy that to the Computer, Omnipage it, and my 300MB of photos are cropped, tidied up, enhanced, then OCR'd into a word doc (around 2mb) or a PDF (around 6mb) - adding pictures increases this number, obviously.

Just so you know, I archive everything that I read. I add a TOC using headers in MS Word, and I add a page number and link back to the TOC in a header that I add to every page (OK, Word adds it) making everything I read much easier to navigate for later reference.

Enjoy.

I tried all of the above suggested methods, & the readme.txt, to to try to get it to register. None of them worked when I re-started the program. Waste of time. Uninstalled. Win 7 x64 as administrator, installed on D: drive.

Little Help? OCR software for pdf scanned documents doesn't make any sense to me. Portable Document File is electronic to be read on the computer. When it gets printed out at some time and is physical and then is read by a scanner using OCR, what is special about the document being PDF?--or actually "used to be" PDF until it got printed to paper?

I have googled for definitions but everything assumes the reader knows why this designation of ocr is required.

I have not yet found a pdf CONVERTER that will print what you see. That would seem to be easy to do to me--all the commands for spacing and what not are in the pdf coding right? Including math formulas and even pictures. The whole thing is just one's and zeros.

Any easy explanation for what I'm not getting?

Seems to have trouble with color changes. A green highlighted row in an Excel spreadsheet was skipped. This will deter me from using very frequently.

@51 - Mike - before I retired I worked in the Trust Pension department of a large bank. I, too, needed to find a program which would do what you have described. It's now been 20 (oh, God, that sounds impossible) years since then, and I guess there must not be one out there even now or you wouldn't be asking for it.

I solved my problem by "drawing" the form in Excel for the user to bring up on his screen and fill out. Then I composed macros in another part of the spreadsheet which would "interpret" their input to an Excel spreadsheet that I could easily use. I know macros are probably a thing of the past, but if they're not, you might want to try this process. If I had owned a spreadsheet program, you bet I would have soon become rich from selling this service to - well, almost everyone who owns a business. As it was, I spent so much time creating forms for other people in the bank, I couldn't handle my original position. So, I got a new position with a very nice raise in salary, to do this all day long.

Give it a try if you work with spreadsheets at all. It's really so simple to do, and so easy to make changes as the need arises.

Hope this helps, AP

Don't waste your time on this. After reading #4- einstein's comment, I almost skipped it. Now I wish I had. Installed just fine. Activated smoothly. (Win7, 64-bit) Worthless in actual use. I fed it actual text-content but SECURED pdfs—it goes into an endless loop with a rotating virgule. OK, it said scanned pdfs. So, I tried several, both single-page and multi-page. The best output result I got out of all of them looked like this (original below, OCRed output on top), which is clearly unacceptable. I'm going to take einstein's advice and give calibre a shot. Hey, it sure can't do any worse...

For 1 paged PDF, converting is in progress 29 minutes now and still working...
Why this?

40 minutes still not working or registering on win 7 64 bit ultimate 2thumbs down un-installing as we speak goodbye have a nice day thanks

Very disappointed! Installed on a Windows 7 - 64 bit platform. Previous downloads from GOTD were greatly appreciated and installed seamlessly. Cannot get pdfocr to register - even tried rebooting. I am deleting the program from the system.

Start the program atleast once as Administrator and enter the registration code...

Or the program will always say 'Unregistered'

Thanks OldScotty, on behalf of many of us here -- your advice is usually useful.

However, after reading all the negative comments and complaints here, my feeling is -- why bother?

I for one have better things to do than to nurse errant programmes through dodgy installation proceedures -- that often do not seem to work.

"Free" then doesn't mean a thing -- it COSTS you a lot of time and effort (and often frustration). I can't imagine too many of us will want to bother to install this programme, given the "reputation" established in the past 40 comments.... Especially when there seem to be other, really free, alternatives.

Sorry I can't comment specifically -- I'll be giving this one a miss.

I am looking for a reasonably priced PDF program that can create interactive forms for people to fill out and then save to their computer. Many programs I have checked out, such as this program, manipulate and create the files, but do not offer the ability to create forms with fillable fields. If anyone knows of such a program, I'd love to hear about it.

My test PDF was a very clear desktop publishing document tranformed into PDF. It should have been read excellently by this software. However, the character recognition in several sections was extremely poor. Bold and italic fonts caused significant problems. Some very large fonts were not read at all.

When proofreading OCR'd documents, I can grudgingly accept small textual errors that are easily spotted and corrected. OCR'ing numerical data, on the other hand, must be near-perfect as those errors often are difficult, if not impossible, to correct. This product produced too many unacceptable issues with digits, which proved to be very counter-productive.

Findings:


From a PDF created with a PDF print utility (CutePDF). Arial and Courier fonts.

Misread characters - a - b - c - d - g - i - j - l - m - o - r - w - ; - : - , 1 - 5 - 6 - 7 - 8 - 9 - 0 - space

Italics - often misreads "o", "i", and "l" (issue worsens with bold italic)

Bold - Misreads often

Large fonts (14+ points) - misreads increase

Small fonts - (7- point) - misreads increase

Numerics - Generally good, sometimes substitutes 1-l, 5-S, 6-b, 7->, 8-B, 9-g, 0-O.

Columns - does not accommodate

Graphics - occasionally interprets pieces of non-textual images as text.

Occasionally, a poorly-processed character will cause an entire print line to be misread.

Wow! Totally impressed. I just downloaded and installed Abby Fine Reader v.11 trial yesterday to translate some images of French text in PDF format. That program did fine but at 767 Megabytes I never figured PDF OCR here stood a chance especially with French. However it DID have the option for French and did just as well as the huge Abby fine reader!!! All in under 15 Megabytes. Anyone complaining about it not doing pictures doesn't understand what OCR is about. It's about reading text in images. So it saves in Wordpad by default with is way faster to load up than Word or most other programs. It's a winner for my purposes.

On my way out of this page, I noticed a freeware program also offered by PDFZilla. It is in the programs shown below the free one for today.

The text is as follows: "There are a lot of PDF files can not be printed or converted, that means those PDF files have been protected. PDF Password Remover is a freeware tool to remove PDF restriction from protected PDF files."

I used the link to check on the site for this software, and it seemed to be something that I might need - or better said, have already needed. If I understand this correctly, we would need such a program before we could use today's GOTD in quite a few instances.

Would someone with more computer savvy than I please check this one out and let us know what you think.

Thanks, AP

@2, @4, @36, @37
This is not a PDF converter. This is an Optical Character Recognition program. I use PDFZilla (2.9) and Calibre myself. They do not contain the OCR functionality.
--------------
This program will perform poorly on text based PDFs. For them please use your reader program or Calibre.

I have several image based PDFs with old German texts. That is the type of German most young Germans no longer learn to read due to differences in spelling, grammar and font.

The result of a test run with one of these PDFs has been a readable text. I then scanned the same PDF with ReadIris Pro 11 ($129) which produced an output that can most accurately be described as "a random collection of letters".

My personal recommendation here is get and keep this software as in some cases at least it will produce much better results than the competition.

Based on the above comments, I deleted the program right after download, before I installed. Thought this was going to be a good thing, being FoxZilla. No time for all the problems this is causing.

@23 - Giovanni. Once more you have saved the day. The online program you gifted us with is really what every PDF converter should be. And I noticed that they offer a more "beefed up" program (not free) that should cover anything not offered in the free one.

Sorry, PDFZilla, but even with my really simple PDF files faxed to me on my printer, your program was a disaster. My document was 2 pages long, no graphs, columns, etc. And it took me longer to edit the errors on your conversion, than it did to simply retype the edited document into my Word program. This one was uninstalled almost before it knew I had installed it. Also, I had the same problem most are having here - unless I pay for it, I can only download 3 pages. Bad, bad idea if you're really ready to sell this thing. I don't want it, and I got it for free.

And BTW, where is the developer to answer some of the questions asked in at least half of the comments? I suppose it MIGHT be possible that there are choices in the software that would help with the problems we are having, but with no help from the dev, no positive feedback from us.

Thanks GOTD and PDFZilla, but I'm sticking with Giovanni for my free apps for PDF editing. He has never let me down with any of his suggestions.

Who can install on XP?

Thanks for the GOTD. Installation and registration worked fine, but the program can not handle foreign characters. (I come from Sweden and it does NOT work for my language.)

I did much better and faster job copying and pasting from PDF to DOC editor, then let this software do it for you.
The TXT file is disastrous using this software, it will take me hours to correct the mistakes and formatting.

Uninstalled.

I'm not sure how to rewrite this phrase, but it is incorrect as is: "convert scanned PDF paper books and documents" What is a scanned PDF paper book? It's an oxymoron. It can be a PDF of a book, it can be a PDF document, but it cannot be made of paper. It will be made of electrons and protons and neutrons, but not paper. English is such a difficult language; a better product description will help to sell a product, I suggest you work on yours. There is no such thing as a paper PDF. Logically, it could not then scan a PDF metal book, or PDF plastic book...

A very useful program. Installed easily. Does what it claims . Tables are OCRed as text. Like it. Thanks GOTD and PDFZILLA.

For those of you who have entered their registration code and yet the program still claims it is unregistered, it is in all probability that it has not been closed again.
So close it and reopen it.

If the worst comes to the worst close your computer down and then reopen it should be registered then..

This is one of the poorest PDF OCR software I ever tried. Way too expensive, poor GUI, Install in the root directory, and the list just goes on and on

Installed and then entered the registration code in the splash panel, it said it was successful and I needed to restart the program. I did and the program came up in an un-registered state. I went through the "registration" cycle a few times to no avail.

I then remembered that upon install I chose a directory to install the program in (program files) rather than the default c: root directory. Sometimes programs have a hard time finding themselves when directory services aren't well integrated into the installer. I uninstalled the program and then re-installed using the default location for the installation and the registration process worked ok. (I didn't like the fact that a few file folders were created on my root directory (pdfOCR and PDFOCR_Output))

I then loaded a 60 page pdf document (with graphics) and started the OCR conversion. The resultant .txt file exhibited a very unsuccessful conversion rate. I could make out some of the text since I knew what the document subject, but the output is not useful. I suspect that the graphics gives the program too much trouble.

I then loaded a 22 page .pdf document that had less graphics and converted it. The program did an ok job in converting text blocks that were all text. I did have some problems when it encountered non-text items, graphics, HTML, graphs, tables, etc.

I don't find this program useful and will uninstall it.

I had installed a previous GOTD offering: PDF to Word Converter that is more useful to me, Thanks anyway.

This is a pretty descent program. Tried with a couple PDF to text and getting most of the data, but getting some garbled text also. Would be nice for it to keep the layout when converting as the same of the PDF. Other than that, nice program! - Thank you!

What a waste of time!. Some of you guys appear to be luckier than me. Windows 7 64bit. You install and a pop-up window appears verifying software with Giveaway of the day. Fine. But it froze and I had to restart my PC 3 times. Please get your act together GOTD.

This is the most screwed up installation I've seen.

It installs MSCOMCTL.OCX in the C:\Program Files (x86)\pdfOCR folder yet mine is already installed in the C:\windows\system and SysWOW64 directories. Same with Richtx32.ocx

Trying to convert a scanned .jpg and after 15 minutes I closed it out.

I located 2 pdf files in the C:/tmp folder and one in the C:/jpgtmp folder. It was named 0.pdf, not the name I typed in. In the C:/tmp were a tmp.bmp, tmp1.pdf, and tmpx.pdf

PDF toText resulted in a file that need extensive editting. Wordpad.exe is installed in the C:\Program Files (x86)\pdfOCR directory for editting purposes.

I'm uninstalling

It would be nice if I could install it: all it does (on initial click) is say "cannot create destination folder"

I tried this a couple of weeks back on one scanned image showing text from a PDF.
The result was a garbled mess. Useless. Even as a freebie

Tried a scanned PDF (British English). Simple text, very tidy picture... Nonetheless... The quality of OCR is so poor... beneath any critique. So far - Thumbs down. Dear developers, next time please try adding advanced options, like user-assisted learning.

I installed under Windows 7. Unfortunately I ended up in an endless loop of being asked to register, registering, restarting, reregistering, rerestarting ... I am going to uninstall it.

Like #15, I am limited to 3 pages, even though it had said it was registered.
Need to download again? Install again?
XP, SP3.

Bill

# very disappointing that it does not do scanned image to text. well, i have used scanned image to PDF and then scanned PDF to text features to get the job done. and, guess what? the result is pathetic, i have used FreeOCR before to process the same image, and it gave better result! PDFZilla wants $50 for this? good luck.
# very irritating that it installs in C:\ instead of C:\Program Files\ by default.
# when you choose PDF to text, it does not let you choose the destination of the output. duh!
# when you want scanned image to PDF followed by scanned PDF to text, you can't do the 2nd job without exiting the program.
# when you want scanned image to PDF, it does not recognize anything but JPEG by default.
# when you want scanned PDF to text, the output opens in wordpad. can't they just use something built-in what FreeOCR or some other program does?
# the filesize of the program is too large for the jobs it does.

although i have liked PDFZilla's PDF to image, this is worthless.

Tested : a basic 3 page PDF file, no image, no table, no column; just plain French text. After a few minutes and 60% CPU load, still nothing. For me, this PDFOCR does not work.

So I tested again with gsview32. It extracted the 3 pages in less than 1 second. Anyone working with PDF files should have gsview32 in their toolbox.

There isn't many free alternatives that work correctly. Seems the tesseract engine is the best available for free (GUIs: Lime OCR, FreeOCR). There is also the Kadmos plugin in Irfanview, which is complicated to configure properly.

Once I needed to OCR a PDF document. After trying the tools above, I eventually downloaded the tesseract-OCR engine and created a small batch file. It worked as expected after I converted each page to TIF.

PDFOCR looks like it also uses tesseract by judging only on the 'tessdata' folder name. Probably modified/altered ---just a guess.

If you need OCR, then you should try various tools. Results depend heavily on the source file, be it an image (TIF/TIIF seems to be the best format) or a PDF file.

Product review:
If you change your mind and decide to open another kind of file, you must exit the software and reload it. Then, it doesn't have an option to OCR from a scan. And, it takes 2 steps to OCR from an image file: first, from image to PDF, then PDF to text ---elevating the risk of losing information. In their web site, they say there's no need for wordpad; but there is a wordpad.exe in the "C:\Program Files\pdfOCR". Go figure...

I have not installed this yet, but I have great concerns.

This is not a thumbs up or down, but just a comment,

PDF are layered files. If the original document was printed to PDF say word to pdf, or xcel to pdf, then the extraction of text is not technically an OCR. It should be 100% accurate, the question is whether this could OCR a pdf that was restricted permissions for change or printing. If if does it has value.

The major players in OCR can OCR an image that was created in PDF, scan to PDF. They sell their products with a API and license. This would mean that if an industry standard OCR say AABBY was used there is a license fee. So I would doubt this would be any better than a decent acrobat file OCR tool Corel or Adobe or Bluebeam.

I have access to these other tools so will pass, but know that you have to pay for the API license royalty for a good OCR engine and a free tool would probably have limited capacity.

Downloaded the program and was to scan some documents. Had to register as it would only do three pages unless registered. Deleted as I thought it would be registered with Giveaway. So no go for me. Thanks anyway.

I have tried it on my Windows Seven 64 bit and gets no problems but the result is in .txt, not .doc file.

Finally a great and very useful program after crappy software released here over the past few days.

As FREE (online) alternative (so no installation needed unlike this GAOTD), I recommend this amazing online OCR service, which supports not only PDF files as input/output but also GIF, BMP, JPEG, TIFF or PNG as input and DOC, TXT or RTF as output.

Let's give it a whirl to see what it can do it for you for FREE!!

http://free-online-ocr.com/

Enjoy!!

also ....
4) it can't handle text that is formatted into columns
5) it needs a way to select areas of text for conversion

Installed without problem on XP SP3.
Seems to work, though I have only carried out limited testing.
There are a couple of immediate dislikes, mainly about unusual (possibly unsafe) programming practices -

1) it installs to C: drive root if you don't intervene

2) it uses C: drive to save files, with no obvious way to change where output files are delivered
3) why use a green arrow to exit the program? I expected it to start the conversion!

The offer today, PDF OCR, mainly works to convert text in PDF documents to text format, but the program is not great in converting the associated images in the PDF document, thus doesnt faithfully reproduce PDF documents to the respective text/Word format as many people would expect. Other reviewers please tell me if you have had the same experience

Slightly different function but possibly of equal value is the freeware "PDFXChange Viewer". I use PDFXChanger Viewer 2.5 Build 200 (ie choose the version with OCR function from the homepage)
http://www.tracker-software.com/

This program is an excellent PDF reader, which also now lets you convert a PDF to one that is searchable. For example, if you have a scanned page or pages from a scanner, converted to PDF format, this program lets you make that PDF searchable with the 'find' function
"PDFXChange Viewer" doesnt convert the PDF to text format, which PDF OCR does

I have no affiliation to either of the above programs