I have a number of images, (these could be vectorised) that are label classifications. There are only 2 classes, 1 or 255. Each image is from a different date. For each date I also have data that can be in csv or point form.

I wanted to try and use logistic regression to get a future, binary prediction using the images, as the dependent variable and the point data associated with each pixel as the independent variables.

I've done this with one pixel, I need to try and extend the model to a wider area and I'm having trouble visualising how to do this. I had thought of having it all as numerical data and looping through the process for each pixel but this would take a very long time.

I am not too familiar with GIS so I could be missing what's possible, but I had also thought of converting the raster to points and populating the attribute tables for each point by joining tables by geolocation. I could convert this values in to rasters as well using spatial interpolation.

I believe there's also some methods in R or python. Or maybe there's some data fusion method I'm missing. I'm a bit stumped on the matter so I'm looking for ideas on how this might be done.

Maybe someone has experience with a similar problem?

Basically I want to input images and text data/point data in to the model and have my output as a predicted image. Maybe this is impossible in GIS.

closed as too broad by PolyGeo Feb 9 '18 at 21:15

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.