ignore background for segmentation of object

Hi

I have a schematic representation of my image (see attachment) which consists of background (black), an object (dark gray) and a star (light gray). I would like to segment the star from the object with the use of a chan vese segmentation (as no threshold is known). However, this segmentation only segments the object from the black background, as these values differ much more than the object and the star. Is there a way to ignore the background or to 'tell' the segmentation to only apply the segmentation on the gray object? The object is already masked, so the background is zero.

Thanks!

[Corey Silva]

I think you will find the "Color Thresholder" app can do this for you. You can also do it programmatically with our image segmentation functions that utilize a threshold. https://www.mathworks.com/help/images/image-segmentation.html?s_tid=srchtitle

[Image Analyst]

Did you try activecontour() yet?

bw = activecontour(A,mask,method) specifies the active contour method used for segmentation, either 'Chan-Vese' or 'edge'.

If so, post your code so we can fix it.

Answer 1

[Selva Karna]

simply way:

1.apply threshold based

2. after apply Binary , you can use Binare selection use bwlabel, u can get background object so u can easily removed

[Image Analyst]

But you left out the code for the chan vese segmentation, which is the method he wanted to use.

[Selva Karna]

I dont know sir?