I have a problem with OCR, help.

[EMETOYS]

I have problems with the OCR of the application, sometimes images are created inside the paper I am working on. there are times when the images on a paper are "pasted". there are times when the texts move. OCR sometimes does't detect some images, iam using Adobe Acrobat PRO DC 18.9.20044. help, i need to modify my PDF's.

[gary_sc]

Hi Emetoys,

What OS are you using and what release?

How are you creating your PDFs (what scanner, what scanning software are you using for that software and what release). When I ask how are you creating, I mean are you pushing the "scan" button or are you doing any adjustment prior to the scan?

Can you share a screenshot of what you are talking about?

Thanks

[EMETOYS]

Hi Gary 

my operating system is windows 7 professional, and I am not doing any kind of scan before editing the pdf, normally it is done automatically.

This is currently my problem, in the paper where I am editing (images, texts) sometimes the OCR transforms all the separated images into one, causing me to not be able to edit the other images in particular. I would like to learn how to solve it. Thanks.

(where it says "this is an image" on the paper are all the images together)

This happens when i want to move an image. (Before the four mirror images were separated, now the OCR put them together)

it's as if OCR isn't doing things right.

Thanks

[gary_sc · Accepted solution]

OK, there's a couple of things going on there.

1) For one thing I'm not completely sure if your computer is fully compatible with this release of Acrobat. Forgive me but I'm a Mac guy and look at https://helpx.adobe.com/acrobat/system-requirements.html, I'm not fully educated on which release means what. I'll have to let you decide that.

2) mixed image and text is always a bear for any OCR to work with/on. 

3) what do you mean that the "and I am not doing any kind of scan before editing the pdf, normally it is done automatically." Where are these PDFs coming from? It looks like you are trying to work off of images from the internet. If so, that's going to almost be a non-starter from the get go because of the very low resolution of web based imagery. 

4) Also, it seems that you are trying to do image & text editing from the PDF. It also seems that the editing you are trying to do is as if you were using a document program like Word or InDesign. That's not what Acrobat was designed to do. While you can do bits here and there of editing (like change the date or spell correct someone's name), the wholesale movement of images is just not what it was designed to do.

Can you please let me know a bit more of where these documents came from and what you are trying to do? I fear that you cannot do what you hope to do here. The best route is to find the original documents and work with them in the original application.

[rodrigoCruz]

Hola, Yo tengo el mismo problema. Trabajo con la versión CC todo el tiempo y sucede que cuando decido utilizar la herramienta de "edición" el complemento OCR se activa de manera automática por lo que no es necesario ir a buscar la herramienta para activar el reconocimiento del carácter.

En concreto, yo estoy trabajando sobre un documento el cual se originó en Word, luego se escaneó como PDF, por tanto viene una imagen en ese documento. Lo que normalmente ocurre es que, Adobe Acrobat, aplica automáticamente el OCR a esa imagen. Creo que a eso se refería el texto de quien hizo la pregunta y que en lo concreto también me sucede. Da la impresión que se hubiera desactivado algún "pluggin". 

Gracias

[gary_sc]

Hi Rodrigocruz,

Let me make sure I'm understanding the process here:

-1 Original document in word (OK so far)

-2 then scanned as PDF (does that mean printed and then scanned or first Word and then Exported to a PDF?)

-3 comes as an image in that document (Are you importing this as a file (like you would an image) into a new Word document?)

Oh wait, I think I'm understanding: After scanning, yes, it is an image. The OCR process can either be an automatic process or a selected process (selecting the command to OCR the page). In addition, if you select the Edit tab in Acrobat, the page is automatically OCRed because one cannot edit an image in Acrobat, only the results of the OCR (and vetorization of any images that have vector components).

I hope I've answered your questions, I'm not fully sure.

[rodrigoCruz]

Hola. acá te dejo la advertencia que aparece en pantalla con mi problema de OCR.

Espero que me puedas ayudar. Ya he desinstalado el programa, reseteado el PC y de nuevo reinstalé Adobe Acrobat CC y el problema sigue...

[gary_sc]

Hi Rodtigocruz,

Thanks to Google Translate I was able to see what the message that was given. 

I've never seen any message that says "Entry Point not found" (or "no se encuentra el punto de entrada"). 

The thing I'm having problems understanding is if you have the original Word document, why are you printing it and then scanning it to create the PDF? If you have Acrobat DC (or Acrobat Pro DC), why are you not creating the PDF directly from within Word from the Acrobat Tab?

Here's a screenshot from my Mac:

[rodrigoCruz]

Hi... I just receive that document, I did not create that. I received my email from other guys to work with that composition to create new editing for my design work. It's only to don't have typing again that text its just that. All times work well but sometimes work bad... and I don't know how to fix that even with that notice in my screenshot that I show you.

[rodrigoCruz]

Dear Friend. I have just been able to fix the problem. It was about moving (copying) the files located in the 'IDRS15' folder that is in the "papercapture" folder and pasting them both in the main "plug_ins" folder and also in the "PaperCapture" folder. (https://helpx.adobe.com/es/acrobat/kb/acrobat-could-access-recognition-service.html) ... and Solved! Thank You very much for your patience.