Since the 70s, NASA, NOAA, and the USGS have been operating a series of satellites designed to look at vegetation health around the world. These satellites, going under the name Landsat, use specialized camera filters that look at light reflecting off chlorophyll to gauge the health of forests, plains, oceans, and even farms. It’s all very interesting technology, and a few very cool people want to put one of these near infrared cameras in the hands of everyone.
The basic idea behind gauging the health of plants from orbit, or the Normalized Difference Vegetation Index, is actually pretty simple: absorb red and blue light (thus our verdant forests), and reflect nearly all infrared light. By removing the IR filter from a digital camera and adding a ‘superblue’ filter, the NDVI can be calculated with just a little bit of image processing.
The folks behind this have put up a Kickstarter with rewards including a modified webcam, a custom point and shoot camera, and a very low-cost source of one of these superblue filters. Just the thing to see how your garden grows or how efficiently you can kill a houseplant.
You can also just use some processed film and pretty much do the same thing, from what I can tell.
After removing the IR filter
Well, obviously
That’s pretty much what they are doing, except that they use a special filter which lets red, green and NIR through. Unexposed processed film blocks all light, except for NIR.
You need to know at least the red and the NIR reflectance to know how well a plant is doing. Plants use red and blue light for photosynthesis. They do not use green and NIR. So the difference between the reflected NIR and red light is a measure of the amount of photosynthesis going on.
A drone equiped with one of these cameras could be very handy for small farmers and environmental research.
You are partially correct on your explanation. What this system allows you to look at is the raw chlorophyll fluorescence, which may give you a rough idea to how ‘healthy/happy’ a plant is. In reality, this basic measurement does not give enough information to indicate plant/photosynthetic health, just if there is chlorophyll present. Granted the difference between 2 images separated by time may tell you about the change in chlorophyll concentration over time which could indicate photosynthetic reaction center damage, but without other measurements it would be supposition at best. It is a really cool project though!!
“[..] or how efficiently you can kill a houseplant.” – This applies to me in so many ways!
Infrared Quadracopter Cameras… :)
Let us be precise. This is a Near-infrared camera with visible light image fusion. It’s a good trick which has been used for military applications e.g. http://gizmodo.com/5933394/nightlighter-optics-find-ieds-hiding-in-the-dark
The important thing to understand is that these systems won’t perform like a thermal camera in terms of being able to pick up a warm body in a cool environment. With carefully chosen sCOMS sensors, optics and really good image processing one can do a lot but nothing like the Walrus thermal camera or a FLIR system.
BUT, who cares? This group has identified a significant need, found a cheap solution and can stand to profit from it. A+
It is really good to see this agile team putting near-IR to use in agriculture and post oil disaster survey.
– RObot
*sCMOS derp.
The government has been spying on my fica tree!
the worst part is that they were right there, and never told you when it needed watering. such a waste of government resources ;)
Prime Directive, and all that
The publiclab.org folks have a few cool DIY science hacks. After listening to a talk they had at KC Maker Faire I went home and built a spectrometer using instructions on their site. They also have a bunch of cool projects with aerial mapping using balloons & kites.