Fresh Install OCR Errors

Report · May 08, 2023

Hello, just installed Adobe Acrobat Pro 2020. Trying to OCR some text from pictures. I get an error message: "Unable to locate PaperCapture recognition service. Your installation may be corrupt."

No problem. I follow the Learn More button, and do what it says. Then I get "Acrobat could not perform Text Recognition on this page because:

Unknown error"

Super. So I uninstall, clear the Acrobat folders & reinstall, and get the same sets of errors (see attachments).

Now what?

Report · May 08, 2023

What kind of files are the pictures you have? TIF, JPG, GIF, PNG?

Report · May 08, 2023

Mostly JPEGs, but I have some TIFs as well.

Report · May 08, 2023

It does the same with both.

Report · May 08, 2023

Can you share one or two of these files? If you want, you can also DM me the files if you do not want them to be "out in the wild."

Report · May 08, 2023

Many of them are OVER the size limit for upload, but here is one of the "smaller" ones.

Report · May 08, 2023

Hi, @KewannaPubLib,

Well, I've got good news and bad news. I was able to OCR the page, or should I say half-page, which led me to the bad news: to do this, I opened the file in Photoshop and cropped it into the left page only. Plus, by saving it as a TIF file, I saved myself a bit of hassle in the 2nd part, the OCR.

Here's a tip: if you save a file in the TIF format, when that file is opened in Acrobat, it will automatically PDF it and then automatically OCR the file.

But, now I have to ask, these files are ginormous! They are also set at 600 ppi. (Good for OCR purposes but also making the files very large). How did you scan these? How big is your scanner? I am curious.

Anyhow, my guess is that these files have too much data on each page, causing Acrobat to choke up on the buildup of data. FWIW, the 2nd time I ran this file as is, it gave me the same error message that you got. Then it started the conversion process a 2nd time that gave me a file that just twitched around like crazy as I was trying to work with it. That's when I cropped the image in half and processed that with no issue.

Also, for the record, the original JPG image came in at 35.8 MB. The left page TIF was 205.1 MB, The first iteration of the left page PDF was 13.7 MB, but then I ran it again for Reduced size, and it came in at 703 kb. Fairly good reduction.

If you can, you might want to try and reduce the size of your scanned pages, save as TIF, and run them. Let me know if that process works for you.

Report · May 08, 2023

Hmm...that is both fortunate and unfortunte.

For the record, we are using an A1 scanner (BookEye 5 V1A), and although the newspapers don't quite fill the scanning area, they do take up a good 2/3 of it, and yeas, the pixel counts on them are quite high (in the tens of thousands). I realized that the file sizes were larger than average, but I figured that Adobe Acrobat Pro would be able to handle it, being a PRO piece of software and all, lol.

That said, I actually DID intend on separating the pictures before OC-ing, as each page will eed separated to maintain proper order. I just don't want to lose very much quality wise in the process, as we will be archiving these long-term, and would like them to be as searchable as possible. Not that I think that halving their size would really "ruin" their quality at all, but a larger source will produce a higher quality end product, ideally.

It's equally surprising though, that a JPG was not as easy to work with as a TIF file, seeing as to the lower file size and all. But I shall try a few TIFs when I get some more time, maybe next week (as I only work here on Mondays). If I run into further issues, I will update here.

Thank you for your time!