after rotating image on matlab..then results go on bad:(

Go To


I am intersted with determining branchpoints of retinal images and then registering these images. First of all I am using a algorithm for extracting of vessels from image then skeletization and finding branchpoints. At first operation all things were ok but if I rotate same image by randomize angle, number of branchpoints are very increase. what are your opinions about this problem...

related links of images :

2012-04-03 20:53
by Fatih BORLU
It's hard to say anything without some examples. Have you tried different methods for imrotate function? All the same - yuk 2012-04-03 21:15
how can I upload image or files to my question - Fatih BORLU 2012-04-04 12:06
You can edit your question in the links below it. Above the text box there is a button to add an image - Bill Cheatham 2012-04-05 01:42 - Fatih BORLU 2012-04-05 21:31 - Fatih BORLU 2012-04-05 21:32 - Fatih BORLU 2012-04-05 21:32 - Fatih BORLU 2012-04-05 21:33


imrotate's standard interpolation algorithm is nearest neighbor, that tends to transform your image in a beautiful saw.

If you rotate your image with imrotate, try passing as the third parameter a better method, like bicubic or bilinear, both will be much better.

The syntax is simple:

imrotate(img, degrees, 'bicubic');
2012-04-03 21:21
by Castilho
thanks for your reply. I tried your suggestion but results are same. I want to show images, how can I upload these to here - Fatih BORLU 2012-04-04 19:14
if the results are the same, chances are you have to rethink the way you're doing things: rotating after the skeletonization is really necessary? If you rotated the image before, and just then skeletonized it, you wouldn't have to deal with this proble - Castilho 2012-04-04 23:36
I added some images. Please, is there anybody evaluate them - Fatih BORLU 2012-04-06 19:53