Correct Recognized Text Finds No Suspects Only in Random Files

[foys40507949]

I am using Adobe Acrobat Pro DC 2015.010.20059 on a PC running Windows 7 Professional. I have scanned documents on our work Xerox machine. I use the Enchance PDF tool to run OCR. This works perfectly fine. Afterward, I can search and find text in the document, so it is clear that the OCR process worked. Additionally, if I run Edit PDF, it will recognize text and allow me to type/edit text in the document. Again, clearly demonstrating that the program has recognized the scanned text. However, in about 70% of all cases, when I try to ensure the OCR has recognized text properly by using the "Correct Recognized Text" tool, I get the message "The document does not contain any suspects." I have files that have the exact same format and were produced the exact same way (paper documents scanned on a Xerox machine), but )one file will allow me to correct recognized text and the other won't. I've spent far too long with Adobe's less than robust support (Case 0187502999 and 0187491863). The answer given is that "some files are incompatible." That is an odd answer given that Acrobat has clearly recognized the text, allowing me to search and edit, but won't allow me to correct the recognized text. It seems like a bug in the OCR algorithm to me. I would upload the files here, but I don't see a way to do so.

[heat_wayve]

Can any Adobe Professional elaborate here? I am having the same problem in 2020... I just want to be able to select a word or section, compare it to the original scan, and correct the recognized font/values. For example, right now I am correcting a document with a row of 'MOS' identifiers that are supposed to be 'M05' - I would like to be able to use a tool that will show the original scan for this portion of my document, but unlike the user above, I have a 0% success rate for the correction tool detecting any suspects. I have also tried the 'Make OCR Text Visible' preflight tool, and have never been able to view my 'invisible layer,' as it were - performing as if there was no OCR scan completed at all. I have officially spent entirely too much time on a problem that seems outdated compared to the advances of technology in our day!