I assume something like you suggest would be possible. It's normal to want to be able to take dozens of long-exposure photographs, maybe even a couple/few hundr

I assume something like you suggest would be possible. It's normal to want to be able to take dozens of long-exposure photographs, maybe even a couple/few hundred. Each exposure might be 10 seconds, 30 seconds, or 2+ minutes long, depending on your equipment. I usually have my camera take all the photos, but I could do this through a laptop, and could script the timing, assuming I had a good feed of the times that satellites will be interfering, and that there is enough time in between satellites to take the long-exposure photos. I'm not sure about varying the exposure times, like you maybe implied, closing the shutter when a satellite comes by. Usually the exposure time is chosen as the longest that the combination of your lens and star tracking mount can do without causing star trails or being overwhelmed by ambient light pollution, and is the same for every frame. Shorter exposures don't gather as much data/light as longer exposures, all things being equal. My first thought while working around satellite interference would be to stay out longer to get more long exposures. It might be the case where having different exposures in a set of stacked frames introduces noise or other unwanted artefacts, given that you would be combining dissimilar inputs. I'm not sure how often these satellites would be passing by and visible. The subject being photographed might be something large, like the milky way across most of the sky through a fish-eye lens, and you would need many of those 1-2 minute long-exposure windows without satellite interference for this idea to work. For a smaller subject, like a constellation, it might be the case where you need to also account for the position of the satellites, if there are too many to track based on time only.