Well, I'm going to stick my neck out and say it's impossible to determine, with a degree of accuracy, the difference between an incoming link and a redirect, or to specifically determine whether it was a redirect and not some other kind of request.
As far as the target website is concerned, it receives the same HTTP request whether it is as a result of a redirect or a user clicking a link. In both these cases the HTTP request is coming from the browser.
However, you could perhaps examine the source of the referring URL (if any) to determine if it was in fact a link the user had followed, but this doesn't necessarily mean that it is a redirect. It could be a server-side request with CURL for instance (although you could perhaps filter out non-real-browser user agents - if these are legitimate CURL requests then they should perhaps be setting their own UA). And the referrer could be faked or suppressed by the browser and by CURL. So the result of this test could be unreliable. However, it might be enough to discount regular links and assume the remainder are redirects?