I need help....

OK, I think I understand.....

So, your experiment introduces a number of degrees of freedom as you change the object your are photographing and a number of the settings on your camera. Since you are trying to establish whether the intensity changes are due to sampling variance or the impact of your changes to settings, I would start by limiting the number of degrees of freedom in the experiment, i.e. take a number of pictures of a fixed object and measure the intensity each time in order to establish the standard deviation of the intensity measure. It actually sounds like you may have already done this. Next change one setting on the camera and repeat the experiment and see if the resultant data lies within a 95%* CI of the baseline sample (*or whatever degree of certainty your are trying to establish). You can the proceed to change each setting or, indeed, the object, one at a time, each time repeating your measurements to establish whether the change gives rise to a statistically significant change in the results.

I must point out here that cameras are not really my thing, so I don't know whether my suggested approach is practical or not. Please let me know if not and I'll try to think of something that fits the sampling constraints.

Stewart

Glad that my reply was of some small value to you. I similarly used to be pretty hot on this, but, over the years, you tend to only really remember the stuff that you use once in a while. There is probably a multi-variate approach that would work here, but I would need to think about that for a while. I'm a bear of very little brain, and prefer to strip problems back to first principles

Happy to take on any other challenges - it's fun to test yourself once in a while isn't it?