Acrobat X: Improved OCR accuracy / capabilities?

Anyone know if there is improved OCR accuracy / capabilities in Acrobat X?

i just tried the acrobat x ocr capabilities and it was very disappointing. acrobat x ocr is not very accurate.
also, acrobat x doesn't format or select paragraphs properly, because each line of text is given a paragraph break--which interrupts text flow. that means any copied paragraph text from acrobat x will require a lot of manual reformatting.
any chance acrobat can use the abbyy finereader ocr engine in the future or as a plugin? finereader ocr is virtually perfect in ocr accuracy. finereader also handles paragraph text flow the way it should be.

Similar Messages

Acrobat X and OCR Accuracy Issues

Does anyone know if you can assign a custom dictionary to the OCR process?
I have thousands of scanned pages in PDF format that I need to OCR.
When I OCR the same PDFs using another product, I get much better results. Acrobat does not see the same word in the same way twice (swaping "er" endings for where there are "or" word endings). It also misses words that other engines have picked up.
Does it spell check?
I know it is possible to go back and edit, but I have thousands pages to do.
You might also ask -- so why not just use the other product? It does not compress the PDFs the way Acrobat can -- and a two step solution is not desirable (OCR for one part, then move to Acrobat to optimize).
thank you for your help.

When I attempt to OCR a document with the main language as Japanese, the resulting exported text file does not contain the appropriate unicode characters. There are just periods for the Japanese letter forms. Is there another option that needs to be selected?

Can I improve PDF OCR accuracy and compression ratio by running it through Acrobat X Std or Pro?

I scanned a book and stored the PDF file with minimal OCR and low compression using mp ex navigator v1. Can I use Acrobat X Standard or Acrobat X Pro to take that PDF file and improve both the OCR accuracy percentage AND the compression ratio, and then it in a new PDF file?
I am running Windows 8 64-bit. It may or may not matter, but I have Office 2010.
Message was edited by: computer-girl

I purchased Acrobat Pro XI last month to try to digitize all my text books (Scanned as TIFF and imported to PDF). I've been pretty unimpressed in general. Specifically to your question, it seems that if a page has a photo on it or some complex image, Acrobat won't deskew it. My version updated just today to 11.0.06 and it still has this problem. I think it does deskew when "Apply Adaptive Compression" is selected because that option causes each page to be broken into parts (text/image).
Just incase you haven't tried, on the Optimize Scanned PDF dialoge box, make sure the slider is set to "High Quality" instead of "Small Size", that may help with the image quality. In my tests, I ended up leaving it turned up all the way.
I have finally put together a workflow to deal with the problem. I use Acrobat to extract each page as a TIFF to a separate directory. I use a third party program to deskew and crop the images. I use Acrobat to reassemble them into a PDF. I use Text Recognition/In this File (Searchable Image Exact) to do OCR.
I am pleased with the results though Acrobat tries to OCR a lot of images and I just get a bunch of jumbled invisible text on my images. I don't mind it though cuz it's invisible, but pretty pathetic given the cost of this product.
I'll mention the third party program I use in a post below because I don't know if I'm allowed to post it or not. I just don't want this post to be deleted for mentioning another program.
Hope this helps others.

Acrobat X: Is there an ABBYY FineReader plugin for better OCR accuracy?

Acrobat X's OCR capability is not very good. Based on my testing and comparisons, ABBYY FineReader's OCR capability is much better! Is there a FineReader plugin that can be added to Acrobat X for better OCR accuracy? If not, can this be a possibility in the future?

ABBYY FineReader is often better, but sometimes Acrobat OCR is better. On some texts Acrobat recognizes word boundaries where FineReader returns multiple consecutive words as a single very long word.

Improved (vastly) Filing Capabilities

As it is today, LR has decent workflow capabilities from the point that you have filed images to work with. The issue is that there is not a decent filing system built into the product. Post-filing is mostly nailed, with some enhancements needed to ease of keywording and overall product robustness.
What is needed at the MINIMUM is a way to set the filename and sequence per folder. That way, as you add images to a folder, the filename is correct. This is probably the biggest pain I am having with 1.0.
Also, we should be able to rename when moving any existing filed image to a new folder. This is functionality in the system I have been using for several years.
All of my images are cataloged with a unique naming convention based upon a specific taxonomy. The folder structure mirrors this.
Here is an example of what I'm looking for:
Copy images from card to 2 different folders
Folder #1 = .\Birds\Long-Legged Birds\Boat-billed Heron\
Folder #2 = .\Birds\Long-Legged Birds\Great Blue Heron\
In folder #1 images should be named DBWBBnnnn ("n" = sequence).
In folder #2 images should be names DBWBHnnnn.
For me at least, this is the workflow I would like to see built in:
1. Load card or point to folder
2. Images are placed in a sorting place (I'll use the term "shoot")
3. Select keepers from the shoot (note a shoot could have hundreds of images)
4. Keyword keepers with an easier method (perhaps a floating window of keyword hierarchy with checkboxes?)
5. Right click on a library folder to FILE images to
6. LR knows the naming convention used in that folder and prompts for confirm/edit
7. Images are FILED to the library folder via a move and rename using the above info
This is what I have today and it works well and is fast. The primary benefit I have in LR as-is is the post-filing processing.
I for one would e happy to spend some time with the development team and whiteboard out some ideas.

i just tried the acrobat x ocr capabilities and it was very disappointing. acrobat x ocr is not very accurate.
also, acrobat x doesn't format or select paragraphs properly, because each line of text is given a paragraph break--which interrupts text flow. that means any copied paragraph text from acrobat x will require a lot of manual reformatting.
any chance acrobat can use the abbyy finereader ocr engine in the future or as a plugin? finereader ocr is virtually perfect in ocr accuracy. finereader also handles paragraph text flow the way it should be.

I have paid for Adobe Acrobat XI Pro OCR but it will not recognise a letter in the serial number for me to update my adobe account and any of the programs I have tried using to convert a PDF file on a Mac to a word doc it is converting file with funny sym

I have been supplied info in an email to copy and paste for a new product purchase "Adobe Acrobat XI Pro OCR" but am unsure how and where to copy and paste to. There is also a link below the serial number. I have tried entering serial number into my Adobe ID but it is not recognising one of the letters in the 24 Digit serial number??? Also I have tried other products previously downloaded to convert a 7 page PDF file on my Mac and convert it to a Word doc but everything I have tried is converting the file to display some text correctly but also displays random symbols and fonts in place of the handwritten info filled in on the form... also getting blank pages included instead of the info??? Would appreciate some help... I am older generation and not always tech savvy, and it is doing my head in haha.

Hi Jock,
I've checked your account, and all is well there. Please make sure that you're logging in with the same Adobe ID/password that you used when you signed up.
Then, clear the browser cache, and try logging in directly to https://cloud.acrobat.com.
Please let us know how it goes.
Best,
Sara

How to improve output accuracy

Hi all
I have to buy an analog output card to simulate pressure sensors output signal to a UUT in the range of 0V - 50 mV. The customer requirement says that the simulating card must provide differential isolated output with 0.1mV accuracy and better resolution. So far after all comparisons I have landed in NI PXIe-4322 card. But this would partly match my requirement because the accuracy level is not matching my requirement.I tried to calculate the minimum accuracy that can be obtained using the example mentioned in datasheet and resulted in 1.12mV. So I need to get atlest 0.5mV accuracy. can this be solved using potential divider network in line and will it improve the accuracy. Have any one used this card or similar card and performed the experiment to imrove the accuracy.If yes what all parameters I have to take into consideration. Please anyone help me in resolving this.have any one used this card or similar card and
d

Hi all
I have to buy an analog output card to simulate pressure sensors output signal to a UUT in the range of 0V - 50 mV. The customer requirement says that the simulating card must provide differential isolated output with 0.1mV accuracy and better resolution. So far after all comparisons I have landed in NI PXIe-4322 card. But this would partly match my requirement because the accuracy level is not matching my requirement.I tried to calculate the minimum accuracy that can be obtained using the example mentioned in datasheet and resulted in 1.12mV. So I need to get atlest 0.5mV accuracy. can this be solved using potential divider network in line and will it improve the accuracy. Have any one used this card or similar card and performed the experiment to imrove the accuracy.If yes what all parameters I have to take into consideration. Please anyone help me in resolving this.have any one used this card or similar card and
d

Acrobat V9 Pro OCR can't produce a file

I am trying to perform OCR on a credit card statement. The statement has 3 PDF pages and except for the non-regular header info at the top of the page, everything is in nice columns - five of them. I specify the output file to be an excel spreadsheet. The OCR engine works OK on pages 1,3. It chokes on page 2 with an error that it cannot recognize any table OR sometimes produces this message: Acrobat could not perform recognition (OCR) on this page because: This page contains renderable text.
I tried the technote soln to convert to .tiff , but that did not work (actually the instruction are not clear: do you rerun OCR on the .tif file or the newly created .pdf that was made from the .tif file...no matter, I did both, and both failed)http://kb2.adobe.com/cps/333/333110.html
I have also seperated the .pdf doc into three individual files, and OCR'ed page two with same results.
I took page2.pdf, scanned it (not with Acrobat), at 600DPI, and tried to OCR it again, same results.
The page contains a bar code in the margin-could this be killing the OCR process? I tried to edit out some of the noise but can't figure out how to delete parts of the .pdf doc.
Also, I highlight only the colums, select Document-> OCR Text Recognition -> Recognize text using OCR....and it does its thing, says it generates output document, but....WHERE? It does not ask me where it should be placed, and I have no clue where it sticks it.....
Any help is appreciated....
JOhn
sample is below:

Really unresolved, but OK.

Acrobat 9 Trial OCR Problems

I am testing the application to see how it would help in creating an archive of a handwritten letters from long ago. There may be as many as several hundred documents.
The letters are all dated and need to be placed in chronological order and will need to be scanned into PDF format. If OCR could be used to create parallel, more readable versions, that would be excellent.
To that end I have tried to test OCR on a couple of the letters. It appears to go through the motions but then I can't find the results. There are no error messages. The program indicates that the task is complete. I believe the last message is, "generating output."
Thinking the results were generated as a hidden layer I tried to make it visible. Nothing there.
I have to be missing the point.
Help! :)

OCR in Acrobat doesn't do handwriting very well if at all.
You may want to check in to some of the applications that are dedicated to doing OCR and see if they work any better.

Adobe Acrobat 7 Professional - OCR Function

I have scanned documents that were saved in Adobe Acrobat 7 Professional. I'm trying to do text comparisons. I try to convert them to searchable text using OCR but it's telling me that it's below the minimum of 144 dpi. Any suggestions as to how I can get these scanned documents to compare for textual differences? I no longer have access to the original (prior to conversion to pdf files) documents. I have at least 10 documents, each over 25 pages that need to be compared for text differences. Thanks!

Save the pages in Acrobat to the TIFF format, using a resolution of 300 dpi
and the color set to monochrome. Then use the "Create PDF from multiple
files" feature to put the TIFF images back into a single PDF to be OCR'd.

Acrobat X Standard: OCR don't work. Program stops. Reinstallation twice wihout success.

Hallo,
da ich über eine englischsprachige Ausgabe von Adobe reingekommen bin, habe ich meine "Frage" aif Englisch gestellt.
Ich nutze Acrobat X Standard. Seit einiger Zeit bricht Acrobat jeden Texterkennungsvorgang nach ca. 30 Seiten mit der Fehlermeldung ab: "Adobe Acrobat funktioniert nicht mehr. .... Sie werden benachrichtigt, wenn eine Lösung verfügbar ist". Gleichzeitig ist auch die Funktion " Deaktivieren" nur nach Ausführen von "Acrobat-Installation reparieren" verfügbar. Deaktivierung und anschließende Neuinstallation in Verbindung mit Rebootimg etc. war niocht erfolgreich.
Ich nutze Windows 8.1. Über nützliche Ratschläge würde ich mich sehr freuen.

I'm scanning for malware now. The "slow email" was not described well, I meant that the "send" command in acrobat is particularly slow. My main issues are the sudden lack of OCR capability within Acrobat and the scanner issues.
I'll see what happens with the malware scan and in the meantime any help would me much appreciated.

Acrobat X and OCR/CJK Support

Does Acrobat X support CJK languages when running OCR? If not, how much is the upgrade to support these languages? Thanks in advance.

When I attempt to OCR a document with the main language as Japanese, the resulting exported text file does not contain the appropriate unicode characters. There are just periods for the Japanese letter forms. Is there another option that needs to be selected?

Acrobat 8.2 OCR problem

We are having a problem with running the OCR process on scanned PDF files. The process seems to run and complete properly, but when the new version of the document is presented on the screen it has been compressed into the bottom left corner of the page. Below is a sample image from one of my scans.

OK, just what are the steps you are taking. Have you only scanned the image directly in Acrobat (no OCR)? Then you are applying OCR. Did you save the file before the OCR and check the page size of that original save? There are just too many unknowns for use at this point.

Acrobat X: improved functions from VBA?

Hello,
I use Acrobat to print create some PDF files from Excel 2010, from Visual Basic for Applications (VBA).
Are there any new and improved functions in VBA for Acrobat X? What is the best/fastest way to print to PDF from VBA in Excel 2010 for Acrobat X?
I currently use something like:
Selection.PrintOut Copies:=1, _
                    ActivePrinter:="Adobe PDF", _
                    Collate:=True, _
                    PrToFileName:=myfileA
Set PDFDistillerApplication = New PdfDistiller
Result = PDFDistillerApplication.FileToPDF(myfileA, myfileB, "")
Is there a better way?
Thanks for any help.

Basically I'm just filling a PDF form by getting it's fields and setting their values by the objects/methods below:
Set gApp = CreateObject("AcroExch.app")
Set avDoc = CreateObject("AcroExch.AVDoc")
Set theForm = CreateObject("AcroExch.PDDoc")
theForm.Open (folder & fileDef)
Set jso = theForm.GetJSObject
jso.getField("10").Value
Again, thanks for any advice.

Adobe Acrobat Standard 8 OCR conversion spellcheck

I have a user who is using Adobe Acrobat 8 Standard to convert a PDF into a editable Word document using OCR. She would like to know if there is a way for Adobe to proof this documents as it converts into Word. For example she previously had a program called OmniPage Pro 12, she would open a PDF from withing OmniPage Pro and convert this PDF into a Word document. As the OCR ran it would bring up a dialog box to correct misspelled words (typically mis-read by the OCR recognition)
FYI, this document was scanned by someone else and emailed to her, so she does not have the original hard copy.

The spell check in Acrobat only works on notes and such. Document text is not checked. She will need to do the check in WORD.

Acrobat X: Improved OCR accuracy / capabilities?

Similar Messages

Maybe you are looking for