Hi Everyone, I am looking forward to being a part of this forum!I would like to use my uav for monitoring some vegetation plots.Does anyone know of an inexpensive camera solution that will take near infra red and visible band images?ThanksPeter
You need to be a member of diydrones to add comments!
i have read many negative reviews about the removing uv film technique, but with the previously mention ENDVI and the info on the above website i am starting to think it isn't terribl
Looking at the NDVI image supplied by ned, shouldn't the water not have any reflectance of these wavelengths, and maybe even absorption? thus you would not see the reflection of the trees in the water in an NDVI image?
Thanks for all your replies, I appreciate your time.
I probably should have mentioned that I am looking to create NDVI for the vegetation plots, but many of you anticipated that one :-).
the two camera system seems to be the way to go for me but I believe that I need to do a lot more reading:-).
I guess I’m one of the deluded. I use a dual-camera system for capturing 4-band images and I do create “reliable” NDVI images. I suggest you look at the Public Laboratory for Open Technology and Science (PLOTS) web site http://publiclaboratory.org to see the work people there have done. Check out research notes related to NIR. I wrote some ImageJ/Fiji plugins to merge visible (unmodified) and NIR images to create false color and NDVI images. It does a good job automating the image-to-image registration and creating assorted image products. Information about that software can be found on the PLOTS website if you search on “plugin”. I attached a graphic a colleague put together illustrating the output from the software.
The NDVI image created using the methods I use have been using a good for relative NDVI since they are not calibrated. I’m working on simple calibration methods. There are more involved calibration methods explained in the peer-reviewed literature that can be followed if you need to do accurate comparisons between different camera systems. The bottom line is that there is useful information about vegetation that can be generated using a two-camera system.
Some folks at PLOTS are working on a system for capturing 4-band images on a small package. Search for Pi-in-the-sky so you don’t need two separate cameras and all the processing is done on-board the system.
Lastly, the folks at http://www.lifepixel.com/ have a super blue filter that passes only blue and NIR and that can be used to capture NDVI-like imagery. The spectral response of blue and red light is quite similar with green vegetation.
The method of removing the "hot mirror" is in my opinion only good for getting qualitative data. Remember the detector is RGB and you are going to need to decided what data to discard or combine i.e. only take the Green channel average out all three channels etc. A two camera system is next to useless in my opinion (there are a couple of deluded folks that claim it works) for NDVI as there is no way in hell you can guarantee the cameras will trigger at the same time /angle etc and when you start working with the NDVI algorithm you are working with exactly the same pixels.
All digital cameras are sensitive to near infrared but they have a filter (hot mirror) to allow only visible light to the sensor. I've removed the hot mirror from cameras in the past in order to do near infrared photography, depending on the camera it can be pretty easy or really difficult. There are also commercial services like Life Pixel that can remove the hot mirror for you. So you could have one camera for visible light and then another camera for near infrared (which has been modified). There are probably specialized cameras meant for near infrared already but I'd imagine them to be more expensive than a modified camera off the shelf.
Replies
Hi i am interested in making NDVI images as well and have trawled the web trying to find info,
i recently found this website ,
FlightRiot - Open Source Single Camera NDVI – Vegetation Health Mapping
http://flightriot.com/vegetation-mapping-ndvi/
i have read many negative reviews about the removing uv film technique, but with the previously mention ENDVI and the info on the above website i am starting to think it isn't terribl
Looking at the NDVI image supplied by ned, shouldn't the water not have any reflectance of these wavelengths, and maybe even absorption? thus you would not see the reflection of the trees in the water in an NDVI image?
please see at www.maxmax.com
I probably should have mentioned that I am looking to create NDVI for the vegetation plots, but many of you anticipated that one :-).
the two camera system seems to be the way to go for me but I believe that I need to do a lot more reading:-).
Thanks again
Peter
What type of budget do you have?
I guess I’m one of the deluded. I use a dual-camera system for capturing 4-band images and I do create “reliable” NDVI images. I suggest you look at the Public Laboratory for Open Technology and Science (PLOTS) web site http://publiclaboratory.org to see the work people there have done. Check out research notes related to NIR. I wrote some ImageJ/Fiji plugins to merge visible (unmodified) and NIR images to create false color and NDVI images. It does a good job automating the image-to-image registration and creating assorted image products. Information about that software can be found on the PLOTS website if you search on “plugin”. I attached a graphic a colleague put together illustrating the output from the software.
The NDVI image created using the methods I use have been using a good for relative NDVI since they are not calibrated. I’m working on simple calibration methods. There are more involved calibration methods explained in the peer-reviewed literature that can be followed if you need to do accurate comparisons between different camera systems. The bottom line is that there is useful information about vegetation that can be generated using a two-camera system.
Some folks at PLOTS are working on a system for capturing 4-band images on a small package. Search for Pi-in-the-sky so you don’t need two separate cameras and all the processing is done on-board the system.
Lastly, the folks at http://www.lifepixel.com/ have a super blue filter that passes only blue and NIR and that can be used to capture NDVI-like imagery. The spectral response of blue and red light is quite similar with green vegetation.
ndvi.jpg
Peter
The method of removing the "hot mirror" is in my opinion only good for getting qualitative data. Remember the detector is RGB and you are going to need to decided what data to discard or combine i.e. only take the Green channel average out all three channels etc. A two camera system is next to useless in my opinion (there are a couple of deluded folks that claim it works) for NDVI as there is no way in hell you can guarantee the cameras will trigger at the same time /angle etc and when you start working with the NDVI algorithm you are working with exactly the same pixels.
All digital cameras are sensitive to near infrared but they have a filter (hot mirror) to allow only visible light to the sensor. I've removed the hot mirror from cameras in the past in order to do near infrared photography, depending on the camera it can be pretty easy or really difficult. There are also commercial services like Life Pixel that can remove the hot mirror for you. So you could have one camera for visible light and then another camera for near infrared (which has been modified). There are probably specialized cameras meant for near infrared already but I'd imagine them to be more expensive than a modified camera off the shelf.