Dear All,

Thanks for all of the great discussion on my last post about the DVI vs. NDVI vegetation indices. We have done a lot more work in the area and have come to a pretty strong conclusion that, for our purposes and equipment, DVI images really nail exactly what is going on on the ground. We've now seen excellent results over dozens of fields and several different crops. I wrote up some of our most recent results on our website at http://agribotix.com/blog/2014/7/3/agribotix-images-an-x-ray-for-your-field so check it out if you're interested.

One thing I didn't mention in the blog post because I didn't want to get bogged down in the details is that NDVI is very sensitive to moisture. In the side-by-side image of the barley farm, the pivot was watering and was somewhere near the road at the bottom. I believe that the large vertical artifact just to the west of the road is due to that crop being very wet. We also observed this when flying elsewhere where a recently plowed fallow field has a low NDVI signal as it should, but a drier fallow field actually has quite a strong signal. That said, I spent several hours at this field with both georeferenced images pulled up on my iPhone and the intensity of the NDVI signal had essentially zero correlation between what I saw on the ground while the DVI signal was just perfect.

Best,

Daniel