I second Fletch's advice to avoid parsing HTML with regex.

However, another problem you may have is that you are not, strictly speaking, matching, but substituting (with the empty string). E.g.:

c:\@Work\Perl\monks>perl -wMstrict -le
"my $urlresult = 'an Alt Img and another ALT IMG here';
 print qq{string has 'alt img': '$urlresult'};
 ;;
 if ($urlresult =~ s/(alt img)//igm) {
   print qq{string had 'alt img', but no more: '$urlresult'};
   }
"
string has 'alt img': 'an Alt Img and another ALT IMG here'
string had 'alt img', but no more: 'an  and another  here'
[download]

(This example only works because I've removed the  ^ start-of-string anchor.) Please see perlre, perlretut and perlrequick.

If you have more questions, please feel free to ask. Please see How do I post a question effectively?, Short, Self-Contained, Correct Example and How to ask better questions using Test::More and sample data.