This is what the function returns:
http://www.domain.com/wp-content/uploads/2014/09/image.jpg

This is what I need:
wp-content/uploads/2014/09/image.jpg

function catch_first_image()
{
    global $post, $posts;
    $first_img = '';
    ob_start();
    ob_end_clean();
    $output = preg_match_all( '/<img.+src=[\'"]([^\'"]+)[\'"].*>/i', $post->post_content, $matches );
    $first_img = $matches[1][0];

    echo $first_img;
}

ตอบ

$output = preg_match_all( "/<img.*https?:\/\/[^\/]+\/([^\"']+)[\"'].*/i", $post->post_content, $matches );
echo $matches[1][0];

Demo
https://ideone.com/8kKyDR

Explanation:

.* => any character zero or more times
https? => http or https
[^\/]+ => any character other than / at least one time
[\"'] => double or single quote

ตอบ

Use this expression:

~<img.+?src=['"]http://www\.domain\.com/([^'"]+)['"].*?>~i

Demo

https://regex101.com/r/kS6fK0/1

I just literally matched http://www.domain.com outside of your first capture group you reference. Note that I changed delimiters to ~ so we don’t need to escape slashes. You may want to change this to https?://(?:www\.)? to make things more flexible. Don’t forget to escape periods. Also, I made your dot-match-all repetitions lazy to save you from headaches in the future (.+? and .*?).

ตอบ

<img.+?https?://[^\/]+\/\K[^\"']+

Demo

https://stackoverflow.com/questions/25876431/using-preg-match-all-to-return-part-of-images-url-in-wordpress