Hi Dave,

You can try filtering the BDNF image before thresholding to slightly blur the image for the purpose of creating ROIs.  It's under "Process", "Filter."  I usually use the median filter with pixel set to 1 or 2.  Gaussian or Mean filters will also work.  Apply your threshold to the filtered image and most of the noise should be gone.

I would then use the "Analyze Particles" function on the resulting binary image to create the ROIs of the particles.  With this tool, you can create a size restriction to exclude any remaining noise that was picked up in the threshold.  It's likely these pixels are smaller than your objects of interest.  Once you've done these steps, apply the binary image ROIs from the ROI manager back to the original data (you cannot measure on the filtered one so I do all these steps on a duplicate).  

process > find maxima > count. If you first select a ROI that includes the DAPI then you have it. Problably it is also a good idea filtering the image lightly previously. If you still pick up maxima that are not real it is posible to exclude by a gaussian fitting. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0122604

Also you can use Analyze > Colocalization threshold or Analyze > Col 2 or the plugging JaCOP to study localization. But first read hese papers, they will help you to decide with type of colocalization analysis (pearson, manders, ...) will suit you better.

Am J Physiol Cell Physiol. 2011 Apr; 300(4): C723–C742.

Published online 2011 Jan 5. doi: 10.1152/ajpcell.00462.2010

PMCID: PMC3074624

A practical guide to evaluating colocalization in biological microscopy

Kenneth W. Dunn,corresponding author1 Malgorzata M. Kamocka,1 and John H. McDonald2

Journal of Microscopy, Vol. 224, Pt 3 December 2006, pp. 213–232

Received 13 April 2006; accepted 28 June 2006

Blackwell Publishing Ltd TUTORIAL REVIEW

A guided tour into subcellular colocalization analysis in light

microscopy

S. B O LT E * & F. P. C O R D E L I È R E S †