Miniscope Board

General Category => Data Analysis => Topic started by: Joma on March 21, 2018, 06:58:50 PM

Title: pre-processing
Post by: Joma on March 21, 2018, 06:58:50 PM
Is there a way to downsample the video clips before starting the analysis suite?
I was thinking about binning - spatial downsampling - by a factor of 2 in each dimension (averaging across neighboring pixels).
what parameters in the code need to be changed? Bounderies of segments should be adjusted to the new frame dimensions, I guess.

best,

Joe
Title: Re: pre-processing
Post by: Daniel Aharoni on March 23, 2018, 04:25:41 PM
Hi Joma,
We often spatially downsample by 2 when running analysis. I would suggest using the CaImAn package, https://github.com/flatironinstitute/CaImAn, and not the analysis code I have up on GitHub. The CaImAn package has some parameters to do downsampling as well.