3689634683?profile=original

Last summer, with the help of the DIY Drones community, Agribotix set out to discover how drones could be used to help growers make better decisions. Some of you followed the journey through our blog posts, and we are grateful for all the discussion and discourse our results generated. By the end of the summer, we had the pleasure of working with dozens of growers across many states and several countries who leveraged Agribotix drones or image processing on different types of crops.

Over the course of our first full year in operation, we found that many growers are looking for a simple, cloud-based solution to process their images into actionable intelligence. We wanted the participants at DIY Drones to be one of the first to know that Agribotix has opened up our drone data processing system -- Bring Your Own Drone™ (BYOD) -- to anyone flying drones for agriculture.

The service takes images from virtually any drone, stitches them and returns a single view of a field.  If you send us near-IR pictures, we will also return the results with a false-color NDVI image as well as a shapefile that can be imported into virtually any farm management system and used as an aid to precision fertilizer application.

If you are using a 3D Robotics flight controller, you can download our Field Extractor software, which will automate the process of selecting the images for each flight, geotagging them and uploading. 

We have tried to make the process as risk-free as possible; we process your results, return a thumbnail, and you only pay if you like what you see. You can sign up and begin processing immediately.  Use the Discount Code DIYFREEFIELD and your first field will be processed at no cost.

If you are flying drones for agriculture, we hope you will give our service a try and let us know what you think.

E-mail me when people leave their comments –

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

Join diydrones

Comments

  • @Neal.  Thanks for your insights on thermal imaging.  It's something we've been testing out on a low level using the cheap FLIR Lepton but maybe we should move that project to the sideline.  Most of what we know about thermal imaging comes from agronomists that we work with in central Illinois.  They tell us that thermal imagery indicates crop stresses about two weeks prior to when they will see it in NDVI.  And they pay quite a large premium, relative to NDVI, for the thermal images of their corn.  Of course the camera their service provider uses in their manned aircraft is probably a very high-end FLIR, something we could never afford to put on our drone.  

    We also have some calls from citrus orchards for thermal imagery to indicate what parts of the orchard are prone to freezing.  It costs one of our customers $5000 in fuel cost for a single night of running the big propellers that flush the cold air out of the low spots.  If they had more information they'd only turn on the fans that are needed.

  • Sorry, Tom - thermal imaging for predicting transpiration rate requires constant monitoring and is really uncertain. As incident solar radiation normals are also not really normal (due to variations in humidity, and or cloud cover), one would have to be operating in a climate controlled greenhouse in order to scientifically deduce any realistic relationships between temperature and transpiration. A healthy plant will exhibit variable transpiration rates regardless of ambient (or surface) temperature, just as any living organism does,

  • As I always tell my first year remote sensing students, if you get lost snowboarding in the mountains and you hear a helicopter, pull off all you brightly coloured gore-tex and scatter it on the ground while jumping around (nearly) naked, otherwise you will be invisible to the thermal imaging sensors on board. Likewise for vegetation, leaf temperature varies throughout the day: from warmer than ambient air temp in early morning, to exactly the same as ambient air temp and asphalt, and dirt temp at around 2..5-3.5 hrs past sun rise to cooler during the morning and warmer in the late afternoon. When dealing with thermal imagery, one has to think thermally. Plants are essentially bags of water and therefore heat up and coll down slower than not water filled things - e.g. dirt, tractors, etc.... Is there potential for thermal RS in agriculture - sure.... but the associative costs of the sensors and the lack of coherent interpreters for the imagery really makes this a moot point. Thermal is great if you are looking at heat loss (energy efficiency projects, etc...), spectral is what you want if interested in vegetation health.

  • @Dan.  Thanks for the kind words and best wishes with your (ad)venture.  Yes, we should explore collaborations and compare notes on our FLIR experiences.   Reach me at tom (at) agribotix.com.

  • Hi Tom,

    This is really great. I am actually right in the middle of building something similar, and it sounds like our minds are in exactly the same place.

    I have to say, I am really impressed with the work you guys have done, as well as your openness to discussing your methods. Certainly, there is no lack of demand in the agriculture industry and these solutions are going to be really great for the industry as a whole.

    Let me know if you'd like to collaborate. In particular, I have done some experimenting with a FLIR camera onboard and would be happy to discuss with you.

  • @James.  You hit the nail on the head.  We use Agisoft Photoscan Pro and QGIS, as well as some custom software to process our images.  Our full price list for image processing is on our web site.  We currently have an introductory offer for US$0.21/acre which includes the complete package of stitched raster image, false color, and shapefile.  

    We'd be very interested in hearing about your image detection algorithm, and perhaps licensing it.

  • @Gerald.  We agree that thermal imaging may provide some good insights for agriculture.  In our case we are interested in using leaf temperature as a direct measure of the transpiration rate.  Certainly the use as an irrigation monitor that you mention is exciting.  We looked into using one of the one-pixel IR sensors that you mentioned but didn't get too far down that road.  Currently we are beginning to test the new FLIR Lepton low-resolution thermal imaging camera.  We are just getting the hardware set up and haven't flown it yet.  The camera is only 60x80 pixels but I think it will be good enough to see what we want in a corn field.  We'd love to hear more about your experience with the MLX90621.

  • I am guessing you just use agisoft/ pix4d to stitch photos, and then process it in arcGIS/ QGIS(OS) to produce shapefiles? I was playing around with this last year and managed to get my data into Trimble, John Deere and Farmade software fairly easily. How much do you charge for your service? 

    If you are interested, i have developed an algorithm to specifically detect black grass in wheat fields using false colour hue tranformations and statistical thresholding techniques. Seems to be working very well!

  • Tom, have you experimented with some very low cost sensors to measure ground temperature?  From what I gathered from irrigation experts, they say that this measure together with NDVI is a great indication of how much irrigation is required. This isn't very important for temperate climates, but there are huge plantations here in Brazil where water is one of the main inputs to their production. I always considered that the MLX90621, although it has very low resolution, it probably doesn't need more if you subdivide the area into larger ones. Such farms work with 1-2km long pivoted sprayers, but I believe they can regulate the waterflow at each section.

  • Neal,  Thanks for your comment.  You are correct -- in the raster images the pivot tower tracks are clearly visible.  The image we put in the blog post was a vector shapefile generated from the raster image.  This is the file format needed for variable-rate fertilizer applicators.  In this case we divided the field into 10m squares. although that size is selected by the customer depending on the type of precision ag equipment being used.  The signal level for that square is the average NDVI intensity from the raster image, which is why the tracks are no longer visible.  The end-user also selects the number of levels, presented here as false colors.  While I don't remember exactly how the prescription map for this field went, it will typically be 70 lb/acre of nitrogen for the highest levels down to around 20 lb/acre for the lowest ones/

This reply was deleted.