3D Robotics

Farm drones are hot!

3689532145?profile=originalThis magazine isn't online, but you can get it on your iPad, Kindle or Android tablet via NextIssue. The article is a pretty basic overview, with examples of using a Rite Wing Zephyr and a hexacopter along with a Canon s100 modified with the IR filter removed. But it's notable that drones are getting this kind of attention in agriculture. 

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • 100KM

    Hi LanMark, AgPixel will be a great product and can save a lot of time and effort of a farmer or the plantation management to made the right decision at the right time. I hope the program will be available soonest possible at the price everyone can afford.

    I'm here in the East Malaysia doing some oil palm plantation aerial mapping. So far we have complete about 10000 Ha of land. I also had bought a MaxMax modify Canon S100 and AgPixel seems like just the program I want.

    May be I can capture some aerial image with the camera and test it on your system ? 

  • Incredible information in this thread! @Mark @Kevin, I find your opinions very interesting and enriching. You are doing an amazing work... I'll keep tuned.

    Greetings

    Anderl

  • Mark, as far as I can tell not owning a Canon SX230, is that the most significant advantage of the S100 over the SX230 is that the S100 has a bigger sensor. Bigger sensor generally = better signal to noise ratio. Other advantages of the S100 include a wider maximum aperture (3.1 vs 2.0), shorter minimum focal length (28 mm vs 24 mm), higher true resolution (the resolution is up-scaled in the SX230), and the S100 is smaller and lighter. For many applications, however, the differences may not be significant. I'd love to see a head-to-head comparison between the two models.

  • Dries, the S100 conversion was done by MaxMax.com. It is a special request item. You have to phone them to get it done. The cost last time around was $925 (camera included). They have a method of internal filter substitution that works well.

  • Yup! 

    Will need to be carrying an on-board computer for other things so some post-processing work on board would be pretty good.  Should think that composing a 4 image strip or similar of NDVI data would be good.  There might be added benefits in terms of to sending NDVI only data on the fly, leaving the raw data for later. 

    As a developer I intend to get out there and work with some farmers / research institutes to understand what the requirements are so that I can tailor my products better - simplest of feed back loops!

  • Mark,

    I find your posts on this subject really interesting - you are thinking a lot of the things I am.  I'd be looking to provide in-the-field capabilities to do the kind of things you are talking about - such as using GPS enabled tablets to assess data right then and there.  and then using the UAS system to treat any issues found.

    I thought the article was pretty good, but undershooting on the price quite a lot for a de-risked aerospace quality UAS system that would be appropriate for use in commercial applications like this.

  • FlyingMerf, thanks for the insight! I also have an S100. Did you do the conversion on your own? I asked 4 private companies and all replied “We do not convert the S100 as we think it will have a horrible hot spot like the S90 and S95..”

    Mark, sure, understand your point. Again, for sure there is a good market for easy-to-use and tailored remote sensing software for precision farming. Still, if there would be a demonstration/limited version or some more description available I would be interested to see. All the best!

  • Dries, there is very little difference in the exposure settings of the converted cameras vs unconverted cameras. At least that is the case for the NIR/G/B conversion in the Canon S100, when looking at vegetation. The blue and green bands are not altered by the conversion, and the NIR band benefits from the high NIR reflection from vegetation. We always use the highest possible shutter speed for the cameras, which is 1/2000 in the S100. It produces sharp images at our typical flight speed and altitude, which are 18 m/s and 120 m. On multirotors you can get blur if the props are not balanced or the gimbal is not properly dampened.
  • @Kevin/Mark,

    On AGpixel..seems to be nice software you created and can be of great value to translate remote sensing tasks to UAV operators and consultants! Do you plan to distribute with open-source license? Would love to try it out! On functionality, I think some vector functionality could be useful (calculate statistics on field level) as well as simple calculator which (with the right ground measurements) could allow to convert pixel values or vegetation indices to something of physical meaning (CHL-a  or N content/LAI/FAPAR..) .

    You can have a look to the Multispec software from Purdue University, which has similar functionalities + some other cool stuff.. and its free!! Go Boilermakers!!

    https://engineering.purdue.edu/~biehl/MultiSpec/tutorials.html

    On the NIR converted camera, when putting an extra filter in front, do you have to increase shutter time to be able to capture enough light or was there no notable difference? Would like to play around with it as well, but not when it would create more blur in the images.

    MultiSpec© | Tutorials
    A Freeware Multispectral Image Data Analysis System
  • Actually, multiple polls are showing that the general population has little concern about the use of drones for military or peaceful applications.

    http://http://www.aia-aerospace.org/newsroom/aia_news/poll_demonstr...

This reply was deleted.