PHP Crawl page code, picture path related issues

PHP Crawl page code, picture path problem
I'm now going to start by crawling the code for the connection URL with PHP code. 、
Ob_start ();
$ch = Curl_init ();
curl_setopt ($ch, Curlopt_url, $url);
curl_setopt ($ch, Curlopt_returntransfer, 1);
curl_setopt ($ch, Curlopt_header, 0);
$retrievedhtml =curl_exec ($ch);
Ob_end_clean ();
Curl_close ($ch);

Now there is a situation, such as the URL of the webpage is http://www.aaa.com/bq/
Then the link address of the picture inside the page is

At this point, the full URL of the picture should be http://www.aaa.com/bq/cq/a.jpg
But after I converted and found that the picture does not appear, later found that the path of the picture is actually
Http://www.aaa.com/cq/a.jpg

How to explain this, how should I judge such a path?

------Solution--------------------
According to common sense, it's not/bq/cq/a.jpg.
You take a good look at this path:

/to the root.
The picture path is http://www.aaa.com/cq/a.jpg
You take a few more pictures immediately understand,/Front no point, jump to the root, there is a point relative to the current path don't confuse it.
------Solution--------------------
The shape is relative to the domain name
The shape or the relative URL.

Don't you write the website yourself?
The light of the others?

