2
Replies
Hi, @sadie_6960, to understand what's going on here, you need to understand how Acrobat scans and presents what you see in the final OCR.
To start out with, Acrobat, as an application, cannot scan. Rather, it uses some software called TWAIN to access the scanning software of your scanner, or, in the case of Macs, it will let you access Apple's "Image Capture," which is the worst scanning software I've ever seen. If you are on a Mac, you are much better off using the software that came with your scanner, save it to an easy-to-access folder (Desktop?) in TIF format, and then drop the pages onto your Acrobat icon in the Dock.
(If you see this on the right in Acrobat on a Mac, you are using Image Capture. Cancel that and go use your scanner's software.)
The next issue is the quality of the scan. If you put the paper in the scanner and did the scan without setting any of the options, this is like holding a camera up without framing the subject, checking focus, and pushing the button. Sometimes it works and sometimes it doesn't. A bad scan will provide poor to very poor OCR. I'm going to put a link at the bottom of this with a bunch of recommendations on how to get a good quality scan to get some good quality OCR.
But, similar to the scan, if your original copy is not very good, neither will the OCR, and that's even harder to fix than just getting a better scan. (More information if you need it, just ask for some ideas)
And lastly, how did you have Acrobat process the OCR? You have three options:
1) Searchable Image
Ensures that text is searchable and selectable. This option keeps the original image, deskews it as needed, and places an invisible text layer over it. The selection for Downsample Images in this same dialog box determines whether the image is downsampled and to what extent.
2) Searchable Image (Exact)
Ensures that text is searchable and selectable. This option keeps the original image and places an invisible text layer over it. Recommended for cases requiring maximum fidelity to the original image.
3) Editable Text & Images
Synthesizes a new custom font that closely approximates the original, and preserves the page background using a low-resolution copy.
If you are using #1, it means that the PDF looks exactly as the original document, because it is. However, there is an invisible layer above the original with all of the mistakes you are seeing when you export it to Word. One (other) way to verify this is to copy the text from the PDF and paste it into a new Word document. I strongly suspect you will see the same strange errors.
I suggest you try both #2 and #3 to see if you get better similarity between what's in the PDF and the Word document. You have to test this with YOUR document because the results you get will determine which one to use.
Lastly, as promised, here's a blog I wrote for Adobe a number of years ago. The one thing that is dated is that now, scanning software can "sense" the kind of document you are scanning and make (semi) intelligent guesses as to how to fix it, rather than you making the corrections in the scanning software yourself. But again, since I do not know the nature and condition of the item you are scanning, I cannot make a strong suggestion here.
I've been doing scanning and OCR for about 25-30 years now, and it's better over the years, but still no panacea. I really, really wish all this attention to AI fixing everything would creep into OCR yesterday, but it hasn't happened yet.
Good luck!
AdChoices