Yep. It would be something like this (untested code):

use strict;
use LWP::Simple;
use HTML::LinkExtor;

my $url = 'http://www.example.com/index.html';
my $file = 'index.html';

getstore($url, $file);

my $p = HTML::LinkExtor->new;
$p->parse_file($file);

my $i = '000';
while ($p->links) {
  next unless $_->[0] = 'img';

  shift @$_;
  my %attrs = @$_;

  getstore($attrs{src}, "img$i";
  ++$i;
}
[download]

Actually, thinking about it, that's not quite right as the image URLs that you'll get back in $attrs{src} will be relative to the main page so you might need to munge them a bit to get the absolute URL. Also by parsing the URL you could probably get a better image filename than the 'img00X' names that I'm using.

--
<http://www.dave.org.uk>

European Perl Conference - Sept 22/24 2000, ICA, London
<http://www.yapc.org/Europe/>