"" 44.#21 Anthony

I noticed this back on the 11th, when I installed another giveaway. If you get the message:

This program is blocked by group policy. For more information, contact your system administrator.

You might have installed a local security policy to prevent the CryptoLocker infection as shown here:

http://www.bleepingcomputer.com/virus-removal/cryptolocker-ransomware-information#prevent

This prevents executables from executing from specific locations on your system. They might include:

C:\Users\[User]\AppData\Local\ – for Vista, Win7 & Win8
C:\Documents and Settings\[User]\Application Data\ – for XP
C:\Documents and Settings\[User]\Local Application Data\ – for XP

Typically most of the installers are self-extracting archives, which extract the install program then execute it. If it tries to extract and run from one of thise locations, the security policy will block it and give you that warning message.

The solution is to extracting it manually to another local folder location, and then running the setup or installer program manually. (I use 7-Zip from http://www.7-zip.org/ as the extractor.)

Good Luck.

Comment by Michael — December 18th, 2013 at 10:20 am ""


Thanks Michael !! It was indeed "interference" from the anti-cryptolocker program I'd installed some time back - & as you suggested - an UnZip to a different folder did the trick !!

Very grateful for that tip - I've now been able to upgrade my Waterfox 64 browser!!

For those asking about handwriting recognition, here are just a few inexpensive products (multiple products each, probably not very good):
http://www.visionobjects.com/en/myscript/note-taking-and-forms-applications/myscript-studio/description
http://www.phatware.com/index.php?q=page/products/windows
CNet lists:
http://www.cnet.com/topic-software/handwriting-recognition.html

Additionally, there are numerous options using hardware. Note that searching is different from OCR. I haven't looked at Evernote in a long time, but it would store multiple possibilities for handwriting that it was uncertain off, increasing the probability of successful searches.

It may have been my Microsoft wireless keyboard interfering with pasting earlier, it tends to miss keystrokes.

If you contact Nuance (good prices aren't available over the web), you may get a salesperson who can't offer you the best deals and who doesn't know that OmniPage Ultimate is version 19 and Dragon NaturallySpeaking is version 12.5. Your salesperson also may not know that bundle prices are cheaper. You don't have to buy a bundle, and different bundles are available. Just ask for a bundle of the products that you're interested in. Last I checked, OmniPage was more accurate than ABBYY.

Ghenghis McCann #49 Thanks for the link to the free ABBYY Screenshot Reader. It works very well, but it does require registration of an ABBYY product before it can be used. Fortunately the free Finereader 5.0 PRO
that XP-man (#10) posted here qualifies.

Thanks to Samrt-Soft Ltd and GOTD.

I have just tested this software.

Result: Excellent.

Though I have to do some small edits, most (about 95%) of the PDF article I used to test OCR recognition was correctly read.

Please note that OCR recognition depends on the size of the font in the original document. If it is too small and compact, then recognition is more difficult to accomplish.

The user interface is excellent. This is evident in the EDIT stage after the OCR recognition is done. The screen is split into two parts. The upper portion shows the original document. Bottom portion shows the scanned result. As you scroll through and edit some text characters, the two windows scroll in synch, which makes it easy to view and compare.

There is a spell check too.

You can choose to output to MsWord, Text, RTF, HTML, PDF and XLS.

If you choose to output to Ms Word, the software is intelligent enough to preserve the multi-column newspaper layout of the original document.

If you choose to output to Text, the software is intelligent enough to unwrap the multi-column newspaper layout of the original document; and present it into a continuous text document.

A lot of thought has gone into the design and through-put of this software.

Excellent program and thank you GAOTD!

Tried to install program. Right after installation it crashed. Wants to send error report.

Uninstalled & reinstalled - same problem.

GAOTD ends today...HELP!

#10 XP-man, Thanks for the link to ABBYY but the program installation asks for Serial No. No Go.
Am I missing something? Please provide links only to programs that CAN be installed.

Installed today's giveaway on Windows 7 64bit and registered with no problem. Ran a comparison with Free OCR (www.free-ocr.net) and found that Smart OCR Pro did a superior job of recognizing the scanned image. I then downloaded and installed ABBYY FineReader 5.0, while it is a rather primitive interface, it did correct translate the same image used to test earlier.

Verdict -

SmartOCR Pro is a keeper; however, the price tag is rather out of line as it doesn't offer the ability to scan a PDF (still wouldn't make it worth the price though). Since most user, business or home, wouldn't need to use this on a daily basis it should, probably, be priced somewhere between $39.99 and $69.99

ABBYY FineReader 5.0 is fine if you miss this giveaway and need a quality OCR program for free. A multi-language version is available here - http://forum.raymond.cc/threads/abbyy-finereader-5-0-pro-for-free.35458/ and instructions for obtaining the key are here http://tweakbytes.com/Thread-FREE-Abbyy-FineReader-5-Pro.

Forget Free OCR as it requires extensive editing of the document to return it to the correct formatting.

SteveB

Thank you Smartsoft for the additional information, downloading took some time because the files are big, but it all seems to work fine. :)

Very much slower than ABBYY Pro 5 on an identical double page spread of text in English from a published novel and made about 4 errors whereas ABBYY Pro 5 made 2.

Thank you, GAOTD. Much appreciated. And Thank You, XP-Man. Merry Christmas!

to 54: thank you for this very important information - but it had to be on the beginning of comments list (and in the program description /help file). to include languages and dictionaries into installation package is better solution than such (not very well looking) link.
but my language isnt listed there. so i didnt try this gotd, seeing poor results of non-english recognitions in comments. i have an old abbyy and it does its job well - for my ocassional use upgrade was not worth the money. if yes, i wouldnt pay for an unknown ocr sw twice more then for abbyy upgrade. i dont believe in success of this ocr, if it is more expensive than the "industry standard" sw without showing some obvious "added value". gotd can be a good idea to get users which can switch to paid version later, but it shouldnt be overpriced...
to 7: thank you, giovanni, for your effort to suggest fine free alternatives. this time i didnt try them (not needing them really), but few times your tips did help me much. even when i missed the gotd (or didnt need it in that time) - free sw is available for free also later...

It's not very nice to charge such a sum (199 bucks) for a frontend to free Google's tesseract engine.

@ XP-man and others

As far as I remember there are a lot of other languages than french for ABBYY 5 available. And if ABBYY can convert French there won't be any problems with English missing "accent circonflexe" and other special letters. And as far as I remeber you can change menu's language to English.

Version 5 still is very good. For later Windows you should try to run it in compatibility mode.

If somebody knows a legal source for free later versions we all would be very glad to get the link!!!!!!

By the way if you are buying a scanner or a combination usually a modern OCR software is enclosed.

(The ABBYY 5 link got here after somebody posted that a free download still existed and "somebody" else later on searched for and posted the link with the new pseudo "somebody", you never heard afterwards and perhaps never will again.)

I am impressed with the formatting capabilities of this software. Some competitor's products are sorely lacking in this aspect of the finished product.

That being said, please don't expect miracles from any "general public" OCR software. Most of these products have severe limitations and all are ultimately subject to the various idiosyncracies of the printed word.

Most documents, when created by their authors, were not "optimized for scannning". This simply means the author almost always does not consider character recognition issues when laying out the design. They do not consider the impact on OCR with:

- font selections that are based on latin script character sets
- character shapes (e.g., the short descender on the letter "g" in the 10 point Verdana font GAOTD uses on this page that could easily be misread as the letter "o").
- kerning (e.g., "cl" can easily be misread as a "d" if the kerning is too tight or an inappropriate condensed font is used)
- ligatures (e.g., a contextual ligature - a font developer doesn't like the amount of space between, say, "fl" or "fi" so they tighten that spacing and then define the two letters as one character in their font).
- serifs (those little tails on the letters may be attractive but they can play havoc with OCR software).
- bolding (this often fills in the spaces between letters, which makes it harder for the OCR software to discern the separate characters).
- font matching (most OCR software doesn't actually "know" what fonts are being used. It's just trying to do a "best-match" for each character it encounters, with its own stored definition.)
- tight linespacing, where the ascenders and descenders on two lines of text crash into each other.
- low-resolution (e.g., a lot of PDF's are "web-optimized" at a mere 72 dpi. That doesn't provide very many dots to use to build a legible character in an 8 point font).
- colour of fonts and colour of background (colour in general, and some colour combinations in particular, can be a nightmare for OCR software).
- shading (see "colour" above).
- text in images (only as good as the capture method that created the image in the first place).
- and the list goes on...
This is why you'll see varying reports on this GAOTD site about the results achieved when testing OCR software offerings. Because the source materials used by the various reporters were vastly different.

These are also some of the reasons why these design techniques and fonts are not relied on for corporate forms that need to be scanned and processed. Instead, these corporate forms typically use no colour, no shading, and black barcodes and OCR-A and OCR-B fonts, for form parts where character recognition is the key requirement.

There are some industrial strength (hence, very expensive) character recognition products on the market that use advanced techniques (huge dictionaries and lexicons, word recognition, grammar analysis, near-neighbour analysis, etc.) but I highly doubt I'll ever tune in here and see one of them being offered for free.

Cheers

For another free alternative, if your OCR needs are limited to PDFs, PDF-XChange has an OCR feature. They also have a portable verison that does not require installation.

http://www.tracker-software.com/product/pdf-xchange-viewer

2013-12-18
SmartOCR downloaded and installed fine on Win7x64 and I ran it on some test files that I use to evaluate OCR programs. One test file is a half-page pdf screenshot from online of an 1880s book that is very clear to read. On that half page SmartOCR took 5 minutes to recognize the text and the result had lots of errors. It looks like when it gets started wrong on a line the rest of the line is all wrong too. In some cases it was two lines that it thought were one line and the resulting text was completely unrecognizable. It had real problems with names and locations and I need it for family history, so those are extremely important to me. Even if I ignore the errors, the slowness is a major problem since, even if it would do a full page in 5 minutes, that's more than 8 hours to OCR a 100-page book. Not impressive.

Besides speeding up the program and not having the errors, something I would like to see in this program, or in any pdf reader, if the following. Have a way to be able to go through all pdf's in a folder by just clicking to see the next page of a pdf and at the end of one pdf have it open the next one and start showing those pages. Then have a way to mark the pages I want OCR'd so after going through the whole folder the program would OCR the marked pages.

snowd

@29: Handwriting recognition is only a pipe dream. The Mocavo genealogy company recently announced a break through (http://blog.eogn.com/eastmans_online_genealogy/2013/11/mocavo-announces-a-major-step-forward-in-automated-handwriting-recognition.html) in the field, but it only works with a limited vocabulary and is still a long way from commercialization.

Was trying to leave a suggestion, but I can't get it to work, so I ask.

works with Microsoft Word and Explorer, why not other browsers and other word programs i.e. LibreOffice

Also for the Smart_Soft support :

Above is the original PDF file, which I have used, a german Cookbook. You can verify the results by yourself. Excellent for testing because of the poor quality. The text is on doublepage 4, left upper corner.

Thank you!

http://jmp.sh/b/ooKVrfwi68AKpIEE5MDv

Installation on W7/64 was without problems. Since others commented about the quality of the OCR output, I decided to try it on a clean PDF but one that had varying font sizes and some pictures -- it was in English. I used this PDF since it was from the internet and thus did not have any artifacts that might appear on a locally scanned document. My intent was to test the OCR recognition with no 'garbage' on the document that might negatively affect the results.

The long and the short of it is that the recognition was not good -- better than Free OCR to Word for example -- but not anywhere near good enough to use without rather extensive editing.

SmartOCR would probably be good enough on a simple document but even then the output would probably need editing. It has a ways to go before it reaches its potential and is truly useable.

Thank you very much for the SmartSoft hint with the language dictionaries and the OCR files.!

Now I can change to german with a visible difference.


OCR with changed language to German :

IV n,‘w i rd nu n D r. Oetker's Backpulv er ,,,Bachl in. a n- gc‘w,'I,,I,‘L "" A of 4|S kg Mehl rechnet man ein Päckchen
D r. Orth e r`:>I Backpulver,,Backjn".

OCR with original settings English :

IV o ,. w i rd nu n D r. Oetker’s Hackpulver ,q)lHaekin‘. a n- Hew t’I,,I,‘[ “” Dur u|S kg Melli rechnet man ein PackcLen
Dr. O,Iker`s Hackpulver,,Baekin”.

As another example :

Nuance ScanSoft PDF Converter Professional 8.1

Wie wird nun Dr. Oetker's Baeltpulver „Baskin" an-gewendet? Auf 92 kg Mehl rechnet man ein Päckchen Dr. Oetker's Badepulver „Bacldn".

Again the original text

Wie wird nun Dr. Oetker's Backpulver „Backin" an-gewendet? Auf 1/2 kg Mehl rechnet man ein Päckchen Dr. Oetker's Backpulver „Backin".

Regarding today's offering, it installed and registered well on Win7 64-bit. However, when trying to OCR a 35-page scanned PDF it took almost an hour to get up to only *35%*. This is unacceptable, even without checking out its quality. Unacceptable.

Re: Giovanni's recommended VietOCR, it installed and registered well, but when trying to OCR the same 35-page PDF as above, it generated an "Out of memory" error message after only a couple of minutes. Unacceptable.

Re: XP-Man's recommended ABBYY FineReader 5.0 Pro, this is an excellent software (I know from using three other ABBYY products), but it reads only images and scans of documents, not PDFs. I had to convert my 35-page PDF document to JPG images to try it. Having to convert a PDF to images just to OCR it is an added chore. Using it I was able to generate a Word document, but it was about 50% gibberish. Unacceptable.

To sum up, the best software (but not free) I have used is ABBYY PDF Transformer, which was featured on GotD a few weeks ago, and which I am lucky to have grabbed. It performed admirably on my test PDF, with about 98% accuracy, and included all pictures. Let's hope it comes around again.

Dear users,

You can download additional dictionaries from this link:

http://smartocr.com/downloads/languages/Dictionary/

Each dictionary consists of two separate dictionary files with the same name but with different extensions. You need both files. If for example you require a German dictionary then you should download those two files:

de_DE.aff
de_DE.dic

Same goes for Spanish, Italian, Russian, etc. In order to download a file , right-click on it and select Save Link As. Remember to download both files.
You will need one more file. Let's call it a language file. It can be downloaded from this link:

http://smartocr.com/downloads/languages/OCR_Languages/

Download the language file corresponding to the language that you require. For example, if you need to OCR files in German language, you have to download the file called German.lpck. Do the same thing for all other languages.

The installation procedure is very easy and should be the same for Windows Vista, Windows 7, Windows 8 and Windows 8.1. After you have downloaded all files you just need to copy them to their corresponding folders on your computer.

Please make sure that Smart OCR Pro is not running. Go to My Computer, double click on Local Disk C:, double click on a folder named Program Files(x86), look for a folder called Smart OCR Pro and double click on it. Now you need to look for two separate folders named Dictionary and OCR Languages. Copy the two dictionary files into the folder called Dictionary and then copy the language file in the folder called OCR_Languages.

Launch Smart OCR Pro, go to the Tools menu, select Settings, next go to the Recognition tab, select your preferred language from the drop down menu and click OK. Now Smart OCR Pro can recognize documents in the selected language. You can change between languages when you need to.

You can download and install all dictionaries and languages, but please keep in mind that only one language can be active at a time. You can not process a German and a French document at the same time unfortunately.

This is how you copy and paste a file: Open the location that contains the file you want to copy. Right-click the file, and then click Copy. Open the location where you want to store the copy. Right-click an empty space within the location, and then click Paste. The copy of the original file is now stored in the new location.

Some OCR languages may lack their corresponding dictionaries. Smart OCR Pro will still be able to recognize text in those languages even without the dictionary files.

Best regards,
SmartSoft Support

I tried today's software but it had trouble recognizing my scanner (Canon MX892).

I then tied that old version of ABBYY PRO 5, recommended by XP-Man and it worked accurately scanning a Fine Woodworking article into Word 2010 pictures and all.

#41 I used roboform to fill the form and was surprised to see that it not only recognized the French, but it found a US version. (en.usa)
It all downloaded fine.

Abbyy PDF Transformer is currently a lot cheaper than before at Abbyy's site - at least for Europe/UK.

#40 BlkCrowe. This might be what you're looking for, a free copy of ABBYY Screenshot Reader. The link is a couple of years old, but I've checked and it still works. It's a 154 Mb download though.

http://fr7.abbyy.com/ScreenshotReader/ScreenshotReader_bonus.exe

I stopped using any software based on OCR technology, because it can never convert with 100% accuracy.
By the time you edited, removing strange characters, spell proof it and re-insert the pictures it will consume hours of your time just to OCR few pages.
Best solution is PDF editor or graphics base picture editor if you ever need to insert, delete or modify any scanned document.
OCRs so '90s technology and are on the way out. There are very powerful PDF editors today that treat the scan as text.

#29 John Dalton.
You're asking a lot in attempting to convert handwritten notes, however, ABBYY PRO 5 OCR does have the ability to learn what it describes as decorative fonts.
If your handwriting is good and consistent you may have a very slight chance of teaching it, maybe worth a try.

Personally I have found ABBYY so good that it is my preferred method of converting PDF files to Word documents.
Initially, if the PDF is not in image format I convert it to images using PDFill PDF Tools (Free) and then use ABBYY, the only disadvantage of this method is that links are not preserved.

Is my installion activated/registered? W7Sp1 x64
On about Smart OCR Pro says:

Licensed to: _____________________________ <== empty space (no underscore)

To #17: I have dragon naturally speaking 12.5 also. have you mastered it yet? I got mine on cyber Monday for $77.00 with the headphones. I have not mastered it yet.

#21 Anthony

I noticed this back on the 11th, when I installed another giveaway. If you get the message:

This program is blocked by group policy. For more information, contact your system administrator.

You might have installed a local security policy to prevent the CryptoLocker infection as shown here:

http://www.bleepingcomputer.com/virus-removal/cryptolocker-ransomware-information#prevent

This prevents executables from executing from specific locations on your system. They might include:

C:\Users\[User]\AppData\Local\ - for Vista, Win7 & Win8
C:\Documents and Settings\[User]\Application Data\ - for XP
C:\Documents and Settings\[User]\Local Application Data\ - for XP

Typically most of the installers are self-extracting archives, which extract the install program then execute it. If it tries to extract and run from one of thise locations, the security policy will block it and give you that warning message.

The solution is to extracting it manually to another local folder location, and then running the setup or installer program manually. (I use 7-Zip from http://www.7-zip.org/ as the extractor.)

Good Luck.

So did anyone try freeware app mentioned by #7?

Is the old version of Abbyy really still better than this gaod?

Pls reply. Thanks!!!

To #10: the Abbyy pro 5 is in French and do not have a usa version.

Installed and activated fine on Win7 x86. As I test software in a Virtual Machine, I didn't have anything to scan or any saved images with text included, so I took a screenshot with RegEdit open. (http://i.imgur.com/04zsdv9.jpg) The results were so bad I decided to give it an unfair advantage by just trying this with notepad with the readme.txt file opened. The results were just as disappointing. (http://i.imgur.com/Ng3mrll.jpg) This looks like it has some really advanced features, but if it can't handle the most basic tasks than I don't trust it for anything more demanding. I'll pass on this one.

While on the subject of OCR - Does anyone have a recommendation on a free (preferably portable) utility that will allow you to grab an area of the screen and OCR that without having to first save the screenshot and load the image. The best one I've used is Boxoft's ScreenOCR that was offered on GOTD a few years ago but have since had to rebuild my computer. There is an open source one on SourceForge called Capture2Text, but it is still kludgey at best to use. I've asked the developer of PicPick to include this functionality but have heard nothing back regarding this request.

I downloaded and test it and it works great. I needed an Ocr program. it runs a little slow, but it works for me. I would not pay $199.90 for this program especially when Giovanni given out freebies.

To those who compare it to Abbyy- doesn’t Abbyy only work with PDFs? This at least can use any image in your PC or scanned.

I have used ABBYY 5 and also the later versions for many years.

I can confirm that ABBYY does OCR from images of text.

And very accurately too.

I have a lot of text in French and Italian !!! the result was very very strange . Is there a solution??????

Loaded a long (200 page) PDF. Loaded and started the scan ok but the program crashed out asking I send an error report to the publisher.

I do legal work and long PDF's are part of it. I was hoping for a program robust enough to handle this work but it seems this isn't it.

Perhaps if it were 64 bit can could use all 12 gig on this I7 machine, but it's not.

The program cannot be activated yet, at least for me: neither at the initial splash screen nor in the main window.

I don't use Microsoft anything but win 7 so I' wondering how well this will integrate with Open Office. If it doesn't it should be updated so that it does as millions of us can't afford MS crap!

1.Installed fine on Win 8 64/86. BUT even though I gave it English to recognize, it did a rather poor job! Paperport is also expensive, more than this one I think, but it does a perfect job! What I gave it to recognize was even written in Arial on a screenshot of a webpage, so it can’t even recognize Arial. That’s kinda strange, considering this ap has been released and supposedly sold to others. If so, I assume they didn’t try it first. Disappointing, as I really need a good one.
2. To those who compare it to Abbyy- doesn’t Abbyy only work with PDFs? This at least can use any image in your PC or scanned.
3. Thanks GOTD, & Smart-soft.
Happy Holidays to all! :)

Can't even get as far as trying to activate/register. "Failed to launch the program" repeatedly, despite two downloads. Most GOTD offerings launch fine. Running XP sp3, which should be supported.

I gave this software a try this morning and have to say I like it. I don't need to use OCR often, only for getting access to the text from pdf's sent to me by colleagues. My requirements may not be as high as those of some of those who already have commented.

I had no problems downloading or registering the software. When I opened it a very simple OCR wizard popped up. I loaded an 8-page pdf with tables, footnotes and charts. On my computer it took about 4 minutes to convert the whole document. The results were near perfect, except for one line of one table that didn't match the rest of the formatting. But that was an easy fix.
This program fits my relatively simple needs, though I wouldn't pay the suggested price.

#14 Giovanni.
Your reaction was as expected.

#15 Paul A.
Instead of whingeing share!

#21 Anthony Green.
Sorry I can't help I had no problems in XP, I'm sure someone on the site will assist you, best of luck.

As far as I'm aware ABBYY uses the same OCR engine today as in version 5, however from my experience with version 11 the output choices are more up-to-date.

ABBYY’s OCR is so good even in version 5 that it would be difficult to improve on it, the wheel is pretty old but it's still good!

As far as I'm aware the program in the link I sent covers 100+ languages.

The link I posted was kindly given to me on this site some weeks ago, unfortunately I can't find his name, but thanks again!

Downloaded fine. Registered fine. Tried a simple photocopied document. A lot of mistakes for a $200 program.

Very much interested in this GOTD. I have to ask the group here though, what is the best OCR for converting handwritten text.

I have numerous PDFs of my handwritten notes that I would like to convert and store on my computer.

Looking at this GOTD I am not sure that it would be able to do what I am looking to do.

Thoughts and/or suggestions?

Question:
Does the program only work with scanned documents? Or will it work with files already created? Such as, a downloaded PDF.

This is a really good software if you work on English texts. I tested it by converting to Word a workshop manual scanned pdf file (40 pages) with plenty of illustrations and mixed texts. Quick and rather accurate in my test. Of course you have to check the result- but this is easily done in the very clear interface. Thank you GOTD and Smartsoft LLC for a very nice Christmas gift.

Dear users,

the activation problem was solved. You don't need to redownload the program, just try to register it now.

--------------------
Sincerely,
GOTD team