The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

How to find the black hole in an image?

P

Pixel

Guest
Hi

i have some image where i need to find out the black holes, the prob is we could able to see some light difference in all the images so we cant able to set the threshold value ...

i tried histogram equalization n normalization, but for some images it is working, not for all images - How can i process further ?

For Ex :
1st Image
Image Background value - 40 45 2 RGB
Black hole value - 18 17 0 RGB

2nd Image
Image Background value - 20 16 2 RGB
Black hole value - 14 15 0 RGB
---------------------------------
based on 1st image if i set as below 20 as my threshold value, it working for these kind of similarly images but it goes wrong when we running 2nd image ..

Any help would be helpful ,

Thanks
Pixel
 
T

tranthienvn

Guest
I'm not familiar with this problem! but i read somewhere about it.
First you should your Artificial Neural Network for Pattern Recognition. ( Simple method Canny, sobel,Otsu,.. failed).
And then you can use Morphological Method.
Anyway, you should show out which kind of your image ( aerial, medical,...). There is no method appropriated for every kind of image.
Good luck!
 
P

Pixel

Guest
thanks for your response, its a medical image

http://profile.imageshack.us/user/pixel_india/

You can take a look at my image and suggest some idea i



I'm not familiar with this problem! but i read somewhere about it.
First you should your Artificial Neural Network for Pattern Recognition. ( Simple method Canny, sobel,Otsu,.. failed).
And then you can use Morphological Method.
Anyway, you should show out which kind of your image ( aerial, medical,...). There is no method appropriated for every kind of image.
Good luck!
 
Top