This is a bit of a long shot and probably not within my skill set but, I need to delineate certain features on a map and extract data (preferably automatically). What I am attempting to do is identify weed growth between blueberry bush rows and get a percentage of weeds to bare soil. I know how to go about this by creating shapefiles between the rows however, this is very time consuming and tedious for the large plots I am working with. I have tried to classify the pixel values to differentiate between weeds and bushes but, the values overlap, displaying some weeds values the same as the bushes. Is there a more efficient way to distinguish the area between the blueberry rows from the rows themselves? I have attached three images; one shows a shapefile I manually created between two rows(as a general gist of the idea I am trying to achieve), one is just a zoom for more detail, and the third is an NDVI that has been classified to better show vegetation and bare ground. If anyone has any recommendations it would be greatly appreciated.