Image Segmentation with Bright Spots Inside Darker Spots

I am working with image processing with some noisy and messy data. I have a bunch of larger, dark blobs and inside some of those are small bright blobs. I need to distinguish the find the small bright blobs in the image.

So the first solution I tried is thresholding it to a binary based on brightness, such that the darker blobs are below the threshold and the bright spots are above. The problem? Brightness is different on different parts of the image. At low thresholds and on bright parts of the image the darker blobs are making it over the threshold and being miscategorized as bright spots. At low thresholds on dark spots the bright spots aren't bright enough to make it over the threshold, and are missed entirely.

Does anyone have any possible solutions?

One that I've tried is finding the gradient of the image (meaning the image showed the change in brightness at any given point) so that absolute brightness becomes irrelevant. The problem is the difference between the dark blobs and the background is similar to that between the bright spots and the dark blobs, so the program couldn't distinguish between them.

If I had some way to normalize the brightness across the image thresholding could probably work, but I don't know of any way to do that.

Any help is appreciated. Thanks in advance!

[Image Analyst]

The solution is image processing. That's about all I can say until you upload an image or two. Usually when people ask for image processing advice they provide an image so we're not blindly guessing.