I have a 2D image which contains R G and B values. I think this should be purely based on averaging of pixels and selecting the threshold to decided whether this is day picture or night one.
Can you provide any kind of different ideas to solve this problem?