Averaging Dark Frames to Reduce Residual Noise
A series of ST-7 Dark Frame Residuals are shown here to illustrate the effects of averaging dark frames on residual dark frame noise.
This image is composed of swaths from four different dark frames of 20 minutes duration.
- The swath on the left shows the residual dark noise that results when a SINGLE dark frame is subtracted from a dark frame.
- The next swath shows the residual dark noise that results when a dark frame made by averaging FOUR darks is subtracted from a dark frame.
- The next swath shows the residual dark noise that results when a dark frame made by averaging NINE darks is subtracted from a dark frame.
- The next swath shows the residual dark noise that results when a dark frame made by averaging SIXTEEN darks is subtracted from a single dark frame.
The same columns are shown in each of the four swaths. The appearance of the cosmic ray hits are important to note in this series as well as the general residual noise level.
The use of a dark made up of FOUR dark images has a significant effect on the residual noise level ( when compared to the results obtained when using a single dark ). Using a dark made from NINE dark frames has a slightly better residual noise level. Using the dark made from SIXTEEN darks actually produced an increase in the residual noise level over that obtained from using a NINE dark average.
The process of producing a dark that is a MEDIAN of N dark frames would actually produce better results. This is a computationally intense process and is generally only doen when cosmic ray hits are excessive.
All images and text at this site are protected, copyright 1997 Brad D Wallis and may be used only for personal purposes. These images and text may not be copied or reproduced for any other purposes without the express written consent of the author.
Return to the Wallis Home Page.