I might be off here, but it seems like resolution would be a real issue. The variation you are looking for has a period of 1 year. To get hour accuracy out of

I might be off here, but it seems like resolution would be a real issue. The variation you are looking for has a period of 1 year. To get hour accuracy out of that you'd need to be able to distinguish 356*24 different relative positions of these stars. It seems to me like camera's aren't going to be supplying that kind of data.