Machine Vision

cancel
Showing results for 
Search instead for 
Did you mean: 

Image normalization

Hello all!
 
I'm needing some help in normalizing an image.
 
I'm converting the image into an array, then I use a function to normalize the array and finally I convert the normalized array into an image.
 
The problem is that the final image is not normalized, it's exactly equal to the initial one (comparing the histograms).
 
Any help on this?
Any example VI?
 
Thanks in advance!
0 Kudos
Message 1 of 6
(5,525 Views)

What are you trying to do in your normalizing function?  I think there are a lot of different versions of normalization.

Bruce

Bruce Ammons
Ammons Engineering
Message 2 of 6
(5,518 Views)
You method will work if the max isn't 255, and min isn't 0.
Have you tried IMAQ Equalize yet ?   That will change histogram of your image.

George Zou
http://gtoolbox.yeah.net
George Zou
Message 3 of 6
(5,515 Views)
Hi again!
 
I already got into a solution.
 
All I do is to normalize the image by getting the max intensity value (using the histogram function) and then multiply the image by 255/(max intensity value).
 
I think equalize does not meet my needs, because I need the histogram with a "valley" in order to threshold the image, using that "valley peak" intensity value.
If I equalize the image, the gray values will try to be evenly distributed all over the grayscale, and then it will be harder to know the threshold intensity value.
Message 4 of 6
(5,505 Views)

Idea is correct to normalizing the image. but could you please tell me how did you multiply the image by 255/(Max intensity of pixel)?

0 Kudos
Message 5 of 6
(3,682 Views)

Use IMAQ Histogram to get the max number, then use IMAQ Multiply..

0 Kudos
Message 6 of 6
(3,665 Views)