Quantifying cells with different membrane stains

I am trying to quantify cells that express membrane stains. So far I have used a K-means clustering approach but it has not worked well. The images I work with have blue stain for the nucleus and a red stain or a green stain surrounding the nucleus.I want to sort those blue cells that express either a red or a green stain and be able to quantify them. Image example attached below.

Any help is appreciated.

Answer 1

[Mukul Rao]

Hi,

You can try the L*a*b color space technique, the example discussed is very similar to your use case.