This is a cross-post from the GIS StackExchange because it gained no traction there.

I have 15 multiband rasters for a total of 45 single-band images (5 dates x 3 sensors x 3 bands per sensor for which I need to apply a unique calibration equation. All the equations follow the format Y=A*exp(B*X), and I have a spreadsheet with all the corresponding A & B parameters for each.

Is there a feasible way for me to apply these functions iteratively using ModelBuilder to speed up the process? I am also willing to learn some ArcPy to make it work. Essentially, how would I specify the A * B variables in the calculator, and then tell it to search the spreadsheet to make the right substitutions? I couldn't turn up any forum or support site posts about this.

My current plan is:

- Rename all the rasters to something logical like: "sensor_date.tif"
- Split into temporary single-bands using "Make Raster Tool"
- Feed output into raster calculator and output calibrated singlebands as .tif
- Feed new singlebands into raster calculators to find several different indices (NDVI, NDRE, SAVI, etc.)

could be done in the raster calculator, but use RasterToNumPyArray and use numpy to do the heavy lifting

In the example above, the sequence is bands (3) X and Y direction shape (5, 5) So the above would represent a 3 band raster with a 5x5 extent. If your raster has the bands swapped, (ie 5,5,3) then you may have some swapping to do, or just slice them out individually and process.

An yes, It took about a minute to produce the results, so you could make a list of A, B coefficients and process within a function.