Line detection when redacting content

[TheDharkMark]

I've received a document with some information redacted, however the sender has confirmed that some data was redacted that should not have been.   

They have said that it is a software error where '...the distinction between two lines of text isn't properly detected by Acrobat in a particular page or document owing to the file size or quality of the particular page.'

This doesn't sound right to me because it is not the whole line of text below /above that has been redacted incorrectly, and in some instances, the redaction occurred where no other text is redacted above or below the text.

Is this statement correct?  Will Acrobat really redact random words on a page?

[try67]

The description is pretty vague, but it is possible that wrong information has been redacted if, for example, the original was a page scanned at a poor resolution, and as a result running Text Recognition (OCR) on it was not very successful. In that case some artifacts could be incorrectly identified as text and then later redacted automatically, if the process was not performed carefully. It's a stretch, but it is possible.

[TheDharkMark]

Thanks.  I did think it is a bit unrealistic and I've asked for more detail.  But I agree that it is more that care was not adequately applied.

It's an FOI release, and contains a number of different documents within the one document.  The parts that have been redacted apply to a 3 page letter, so I find it hard to believe that two pages were scanned fine and had redactions applied correctly, but the other page did not.  The redacted pages also don't have any legislative reference applied, where all the others do...