Take a look at HTML::Parser or HTML::TokeParser (which is simply an alternative interface for the former).

Super Search can definitely help you on this one. For starters:

• Re: Remove HTML tags from document (read this one first)
• Re^2: How to remove HTML tags from text