Hello guys,
Hope you are doing well.
I trained a pixel classifier, and it can detect the cells of my interest almost perfectly. However, when I “create object” out of the detection/prediction of my classifier, it objectifies some random pixels where there is nothing. It also sometimes objectifies one cell as two, making it impossible to know how many cells I have.
I feel that this problem is mostly because the pixel classifier function is more sensitive to intensity than anything else (I assume). Do you have any suggestions on how I should fix this?
thank you for your attention I attached the related photos