I am working on the problem of finding a light curve of a supernova for which I have a set of timestamped Astro photographic images, all of which have a different patch of sky.
I am trying to blind-solve those images using
nova.astrometry.net, but most of the pictures I try, fail (>90 percent). I also tried cropping the images having the target nova to eliminate the artifacts (like trees, people, clouds, etc.) but still haven't improved much.
Can someone suggest what to do in that case? I also tried other plate solvers, but none of them are working.