Normalized difference vegetation index, or NDVI, is a method of making healthy vegetation pop in a photo whereas stressed vegetation -- due to water stress, nutrient deficiency, disease, etc -- will not. That information is the kind of actionable intelligence that we want to deliver to the farming community.
The basic theory is that chlorophyl strongly reflects near infrared light (NIR, around 750 nm) while red and blue are absorbed. Of course chlorophyl reflects strongly in the green, which is why leaves look green to us, but the reflection in the NIR is even greater. I'll post a blog soon with a lot more quantitative information about this for folks who are interested.
The basic problem is that the only commercially available cameras small enough to fly on a UAV that record in the NIR are both very expensive and very low resolution. It's probably not a big surprise that hackers take this situation as an affront. In my case, the hacker of choice was Jeff Taylor at Event38. He posted a great blog giving step-by-step instructions showing how to convert a Canon SX260 point-and-shoot into an NDVI camera. I made the conversion over the weekend and was surprised by how easy it was. And voila, a high-resolution camera that senses in the NIR at a reasonable cost. Jeff also sells already-converted cameras for those who don't want the hassle.
Now the easy part is over and the hard part begins -- figuring out how to tweak the white balance and do the post processing to tease the NDVI signal out of my RAW images. I'll post again as I get farther down that track. But fortunately I'm standing on the shoulders of giants -- the Public Lab has been doing some great work in this area and posting it on their blog.