Google treats the text content of PDFs just like normal web pages in most respects. The way they tell something is a PDF is by the HTTP headers the server sends, and just like with normal web pages they discover it with their web crawler based on incoming links.

If the PDF is your own content from the same domain, they will probably be lax about the duplicate content issue. It's common for websites to post things in both HTML and PDF format, especially things that are meant to be printed, like resumes and handouts. Ideally you could set up a meta alternate tag in the HTML page pointed at the PDF version, but even that might be overkill.

If the PDF you are copying is from another domain, chances are that they will notice, and there is a chance they would hit you with a "copycat" penalty. I wouldn't look to the PDF format for any means of skirting Google's copyright/ownership filters.