A short story about automatic exposure with RPiHQ #793
AndreasLMeg
started this conversation in
Show and tell
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
A short story about automatic exposure with RPiHQ (translated by google):
I started with allsky six months ago and the first really annoying thing was the transition between day and night.
Naive as I am, I immediately tried to find a solution. (Step 1: calculate the mean value of the image and try to keep the exposure time at 0.5)
Relatively quickly, however, this brought headaches and sleep deprivation from constant observation of the pictures and logging and the realization that every adjustment also worsened
That was the reason to take a closer look at the keogram. Was not usable for me either, so there were various extensions (Step 2: expand, channelinfo)
Now I was able to observe the mean value over the night and determine anomalies or potential for improvement:
At first I had jumped on the exposure time. I quickly recognized the problem that this was not sufficient for the sunrise / sunset (Step 3: gain regulation)
An attempt was now made to start with minimal gain (1 for day, but only applies to RPiHQ) and to increase the exposure time as desired (60s, and one exposure every 60s). Then the reinforcement is pulled up to the desired one (1,2, ... 15)
keogram feedback: doesn't look too bad, but there were massive jumps at gain 1,2 and 3! -> (Step 4) Research: Relationship exposure time, gain, effective exposure
With this realization (my secret) the results got much better (fewer jumps)
Keogram feedback: Regulating restless or always too late -> (Step 5) Prediction (also my secret) of the next exposure time based on the past
Keogram feedback: Sufficient for me, but still optimizable (I have other things to do besides Allsky ...)
... Maybe a never ending story ...
Configuration:
config.sh:
GUI: activate "Auto-Gain"
Beta Was this translation helpful? Give feedback.
All reactions