I wanted to see if it was possible to pull out proton flux data from SOHO images. I used the Coronal Mass Ejection event (CME) of November 9, 2000.
Here I processed out the background images of the Sun to extract the noise. The noise was the data I wanted.

I downloaded the SOHO movie of this time period, imported into layers in Photoshop. I then duplicated one of the quiet layers to serve as the control background, and set the blend mode to "Difference". I removed any residuals by using a Black and White filter to remove blues (some of the differences came out blue) and increase greens to white pixels. A threshold filter was added for the ultimate contrast-enhancement.

I then selected a non-Sun section of the image (in the left corner) and used the measurement tool to count numbers of hot pixel zones, and the combined area of the hot pixel zones. [I tried using the entire image, but the measurement tool choked due to the high numbers].

A plot against elapsed time in hours, and voila, the rapid birth and slower decay of a passing proton storm over 30 h as evidenced by hot pixel strikes in the camera:

Click to view attachment

Noise ain't so bad sometimes....