It is very difficult to parse HTML using regular expressions. You will be better off using a module that understands HTML better than regexen. I usually use HTML::TreeBuilder.

It is very difficult to parse XML using regular expressions. You will be better off using a module that understands XML better than regexen. I usually use XML::Simple.

use XML::Simple;

my $str = "<story>
                <page>
                        <description> desc </description>
                        <image> notExpected.jpg </image>
                        <headline> head </headline>
                </page>
        </story>
        <image>correctImage.jpg</image>";

my $xml = XMLin( "<wrapper>$str</wrapper>" );

my $image = $xml->{image};
print "$_\n" for ref( $image ) ? @{$image} : $image;
[download]

Notice that XML must within a single tag. Hence the addition of <wrapper>...<wrapper>

You're not using [...] character classes correctly.

• [ABC] matches any one of the letters A, B or C.
• [^ABC] matches one character that isn't an A, B or C.

See perlre, perlretut for more details.

edit: I guess I wasn't properly awake. I didn't notice it was XML rather than HTML.
edit^2: Added code