As a starting point for using ImageJ, I would suggest that you take a look at FIJI (http://fiji.sc/Welcome), the tutorials provided (http://fiji.sc/Tutorials) and the many plugins (http://fiji.sc/Category:Plugins) that come pre-installed.
Good options for what you want to do might be:
http://fiji.sc/Cell_Counter
http://fiji.sc/3D_Objects_Counter
http://fiji.sc/SpotCounter
The best choice will depend on exactly what kind of images you are working with and what kind of data you want to extract from them.
You can use DAPI/Hoechst a counterstain and use the Analyze Particles module of FIJI to automatically count the nuclei, and then automatically count the GFP spots, and divide GFP+ counts by DAPI/Hoechst+ counts.
I give a general description of the Image Cytometry section of the methods in this paper:
Article K+ regulates Ca2+ to drive inflammasome signaling: Dynamic v...
Thanks Svetlana! this is helpful. i have also youtubing on this, the problem i am facing is now that the images that took are on the EVOS microscope means there is also quite bit of dirt on images.. it seems one needs a clean image for counting..
If the debris that is causing an issue from your microscope is smaller than the signal of your sample, you can use a Despeckle command in ImageJ to reduce their size. Additionally, you can size threshold out the debris in the Analyze Particles routine in ImageJ by making your accepted size minimum larger than the largest debris particle. Whatever the case, you will have to report all procedures/parameters associated with your analysis. Just keep track of what you do.
Thanks Jordan! I actually am using image J for first time; I think i can get to where i can count cells/gfp expressing cells etc. If you have some time i can send you an image to try a count.. i understand if you thought otherwise. thanks in any case