Note that there are some explanatory texts on larger screens.

plurals
  1. POImplementing ridge detection
    text
    copied!<p>I'm trying to write a ridge detection algorithm, and all of the sources I've found seem to conflate edge detection with ridge detection. Right now, I've implemented the Canny edge detection algorithm, but it's not what I want: for example, given a single line in the image, it will effectively translate it to a double line of edges (since it will record <em>both</em> sides of the line) - I just want it to read the one line.</p> <p>The <a href="http://en.wikipedia.org/wiki/Ridge_detection" rel="noreferrer">wikipedia article</a> about ridge detection has a bunch of math, but this kind of this doesn't help me as a programmer (not that I'm averse to math, but it's not my field, and I don't understand how to translate their differential equations into code). Is there a good source for actually implementing this? Or, for that matter, is there a good open source implementation?</p> <p>Edit: here's the simple example. We start with a simple line:</p> <p><a href="http://img24.imageshack.us/my.php?image=linez.png" rel="noreferrer">http://img24.imageshack.us/img24/8112/linez.th.png</a><br><br></p> <p>and run the Canny Algorithm to get:</p> <p><a href="http://img12.imageshack.us/my.php?image=canny.png" rel="noreferrer">http://img12.imageshack.us/img12/1317/canny.th.png</a><br><br></p> <p>(you can see that it's thicker here - if you click on the image, you'll see that it really is two adjacent lines with a blank in between)</p> <p>Also, I'm writing in C++, but that shouldn't really matter. But I want to code the <em>algorithm</em>, not just write <code>SomePackage::findRidges()</code> and be done with it.</p>
 

Querying!

 
Guidance

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload