Interesting, I got this exact same question asked last week by a customer. I discussed this with our PDF guru Colin van Oosterhout and he informed me that there is no way to recognize from the 'outside' whether a PDF has been OCRed or not.
However, he *did* mention that using Acrobat batch OCR if you have a directory stuffed with OCRed and non-OCRed files... that Acrobat actually 'skips' the files that have already been OCRed.
I haven't gotten to test this behavior for PDFgen, but it's worth a try. Let me know if you find something interesting?