LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

image segmentation using labview

Solved!
Go to solution

hello, am doing my project in the title called "labview based brain tumour length and area measurement" but while am doing in image segmentation using k-means segmentation i cant get the segmented image so i need  help

0 Kudos
Message 1 of 8
(229 Views)
Solution
Accepted by topic author Nardimes
0 Kudos
Message 2 of 8
(192 Views)

thanks for ur replay but am doing this project in brain tumour detection so k-means segmentation must cluster out the tumour part since it is more brighter than the other side in mr image

 

0 Kudos
Message 3 of 8
(132 Views)

@Nardimes wrote:

thanks for ur replay but am doing this project in brain tumour detection so k-means segmentation must cluster out the tumour part since it is more brighter than the other side in mr image

 


I understand what you are trying to do.

 

You are using the k-means.vi from the MLT in segmen.vi ‏29 KB

it looks like you are having trouble to set it up properly -

the link I posted in my previous post contains a working example using exactly that k-means.vi including example images.

 

How about you post a typical example image of your application?

 

0 Kudos
Message 4 of 8
(97 Views)

am using MR image 

0 Kudos
Message 5 of 8
(84 Views)

how can get segmented image i attached block diagram

0 Kudos
Message 6 of 8
(41 Views)

sir please help me it doesnt display segmented image

0 Kudos
Message 7 of 8
(33 Views)

@Nardimes wrote:

am using MR image 


Hi Nardimes,

you force me to guess what you want to do.

It is pretty likely, that you want to segment the bright pixels - which I suppose to be a tumor (foreground) - from the rest of the image (background)

 

 

If that is the case, this is a non-trivial task to do.

This is because in both images  those tumor-grey-values  occur in more than one region.

1.PNG

 

You need to choose a different segmentation approach,

or introduce more steps in your image processing, e.g.

distinguish between object and no-object pixels,

define a region of interest with-in the ovject pixel-region.....

 

I can not help you with that

 

 

0 Kudos
Message 8 of 8
(16 Views)