Additive binning to create 32 bit image.

[Dvelledge]

I am taking over sampled images with a CMOS chip. I want to do software binning 2x2 but adding pixels instead of averaging. I am starting with a 16bit integer image and I want to end with a 32bit integer image that would basically have a lot of room at the higher end of the histogram. creating a HDR image. It seems as i do this in pixinsight using pixmath. It seems because PixInsight turns everything into a 32bit floating point, its does not seem to give me the results I'm looking for. Any advice would be helpful.

Don

 

[cloudbait]

I am taking over sampled images with a CMOS chip. I want to do software binning 2x2 but adding pixels instead of averaging. I am starting with a 16bit integer image and I want to end with a 32bit integer image that would basically have a lot of room at the higher end of the histogram. creating a HDR image. It seems as i do this in pixinsight using pixmath. It seems because PixInsight turns everything into a 32bit floating point, its does not seem to give me the results I'm looking for. Any advice would be helpful.

Don

There's no real difference between adding and averaging as long as your averaging is performed in a floating point space (as it normally is). The only difference is that the pixel values are scaled differently. Your 32-bit integer output values aren't going to give you anything that floating point values won't. Another way of looking at it is that the dynamic range that can be expressed with a 32-bit floating point value is far in excess of the physical dynamic range possible from a stacked set of camera images.

 

[Dvelledge]

thanks. that makes since.