I found an issue with the test image where the ground range to slant range conversion was not well behaved outside the bounds of the image. I checked some actual MiniRF images and this wasn't the case there. After some failed attempts at monotonic polynomial fitting, I ended up just using a linear conversion function to guarantee it was invertible. This also improved our tests so that we can test within the proper tolerances for groundToImage and imageToGround.
jessemapel
commented
4 years ago 
I found an issue with the test image where the ground range to slant range conversion was not well behaved outside the bounds of the image. I checked some actual MiniRF images and this wasn't the case there. After some failed attempts at monotonic polynomial fitting, I ended up just using a linear conversion function to guarantee it was invertible. This also improved our tests so that we can test within the proper tolerances for groundToImage and imageToGround.