Image analysis of a vitrified latex using Fiji/ImageJ
The specimen was a soft polymer latex embedded in vitreous ice and imaged by cryo-TEM. The objective is to subtract the background (the ice has thickness fluctuations) and then segment each latex particle. We will use shape classifiers (circularity and aspect ratio) to select the isolated particles.
The scripts are written in jython
I'm trying a watershed based on this over-segmentation example.
The watershed segmentation has a problem. I need to figure it out...
Left: background subtracted, Middle: Gaussian Blur, Right: Classic Watershed Segmentation