The window size depends on the application but also on the standard deviation. For larger standard deviations you have to use larger windows. Otherwise, if you have a large standard deviation and a very small window size, than your filter becomes almost a mean filter. However, no need to have the window size larger than 6 times the standard deviation because furthermore the signal is almost 0.
Gaussian Smoothing Window Size and Digital Image Processing
-
30-06-2022 - |
Question
Details on Gaussian Smoothing http://homepages.inf.ed.ac.uk/rbf/HIPR2/gsmooth.htm
I have to implement a image smoother in a project I am working on and I have everything all worked out except window size. I am having a real issue with finding the proper window size. What is a general rule about it?
How do you pick a good window size for your gaussian smoothing algorithm to use?
Solution
Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow