ImageJ and Python based Particle Counter

As a part of my master's thesis work, I have process images and count number of clusters of different sizes. A cluster of size 4 denotes that the cluster has 4 particles. Suppose we have a image of plattinum coated silica particles which has many clusters of particles. Each cluster could of different size. Our goal is to count the number of clusters of size 1, 2, 3, 4, 5, and >5.

Methodology

1. ImageJ is a Java-based image processing program. First, we we use ImageJ to process the image. Then we find the area of each cluster in our image. We export that as a .csv files.
2. Now, we based on the image and areas, we find the area of a single particle by careful eyeballing.
3. Now, we use Pandas, Numpy of Python and read all the .csv we have. Then using the area of the single particle, we find the number of clusters of size size 1, 2, 3, 4, 5, and >5. For example, a cluster of size 3 will have thrice the area of the single particle.
4. Next, we plot all the result.