How can I get total intensity of particles in noise image?

I have a spinning disk confocal image which has a lot of noise and particle(my interests have differnt size and move slowly over time). I need only intensity information of selected positions over time in diffraction-limited images.

I'm using Track mate to detect particle and get their coordinates.

I want to apply 2D Gaussian fitting to particles which have a different offset of noise. Then, I would measure the intensity of each particle.

What and How can I do for that using Image J or Matlab?

Or could you recommend the best way to analyze intensity fluctuation of particles(3D, timelapse)?

Answer 1

[Image Analyst]

You can get in a loop where you measure the mean and SD. Attached is an example where I get the mean - make adaptations as needed.

[Jaehwi Bong]

I'm sorry not to describe precisely. In my image, I need to define my objects first. The diameter of my object(3 dimensional shape) is from 200nm to 1 um(in images of 512*512, pixel size: 0.21um). I want to detect my objects which are being formed and disformed irregularly. And measure the intensity of the objects every time.

1. To detect the particles, I'd like to use 3D Gaussian Fitting which gives offset (It must be noise value below the offset). Then, measure integral density of my ROI(object) without noise value.

I attached one plane of stacks with 50 time points as an example to approach from 2D to 3D.

I wonder how I can apply 2D gaussian fitting to detect my objects.

[Image Analyst]

I haven't looked at the video yet but if you have bright blobs that you want to model as Gaussian peaks, then I'd first see if you can threshold the frame and detect the bases of the blobs. Then ask regionprops for the bounding box. Extract out the subimage of the bounding box and model it to a Gaussian using something like fitnlm or fitgmdist(), in the stats toolbox.