I want to run some time series analysis on sentinel-2 data in EE for my master thesis. However the algorithms for Sen-2 in EE are scarce. The biggest issue is that there isn't any for atmospheric correction. I am extremely new to coding, and while I can handle myself with begginer level tasks, scripting a model-based atmospheric correction method is way out of my league.
So I am looking for a solution that would allow me to stick to EE instead of downloading huge set of images and processing them in sen2cor and in R.
So far I have considered following options:
from what I understoond, there is an option to load your own python libraries to EE (using docker??), there are python libraries for remote sensing including atmospheric correction methods so perhaps it would be possible to use one
scripting in EE code editor an automated version of DOS from Chavez (1988), the DOS correction returns suboptimal results and I would have to figure out how to automatically find the dark object value on a histogram
using this guy script for 6S correction and launching it using docker - this sounds promising but I didn't quite understand how dockers work and what they are - does using them mean the computations are done in EE or on my PC?
launching sen2cor in EE through docker - this is probably not doable on my level of experience, I've seen a topic here where a guy tried to do it and it seems it involves customizing docker environment and sen2cor to digest EE's metadata structure
Do you have any experience with Sentinel-2 in Earth Engine? What would you advice me to do?