Sure you can capture aerials, but no photogrammetry process is going to capture piping/tube steel that thin unless you get 0.25 in/px which is nearly impossible unless you are right up on it and then it would be degraded if the map images from a higher altitude were included You could probably fly it very close and model it by itself… As for capturing the whole property in one picture it depends on how big it is. I estimate about 125-150 acres in full frame within legal flying limits.
I don’t see that the Sequoia is supported, but that usually means that DroneDeploy will be able to control and trigger the camera. Even though, you can fly it manually or with another software and upload to DroneDeploy to analyze the maps and model.
Thanks for the quick response. The biggest piviot I have is 65 hectares(160 Acres). The idea is that I only take one picture per pivot due to the amount of pivot’s I have it would not be linguistically feasible to take multiple images per pivot. How high would you estimate the drone to be flying to capture 150 acres?
Depending on the drone you are going to fly you could get up to 300ft and 70/65% overlaps to maintain decent resolution to be able to see the irrigation and get some relative topography. This with most prosumer grade drones will take 25-30 minutes or 2 batteries to map which should give you just enough juice to get your one overall aerial.
Not sure I understand you are talking about 2 batteries. That would mean two flights to take one single image of one land. I have 80 pivots each around 100 to 165 acres. Now I want to know is it possible to fly to the center of the pivot and take an single images from the center and return or does the software require you to take multiple images of the single pivot.
I think that we need to clarify that this is primarily a mapping forum which means in this case taking upwards of 1,000 images to create a high resolution top-down (orthographic) map. I was understanding that you wanted an oblique aerial from the side like you would see of something like a building and then an approximate setup for an orthographic map the way we create them. If you are wanting to be in the center capturing the orthographic image then it is not possible within legal limits. To capture a 4:3 ratio shaped piece of land at 150 acres you would have to be approximately 1500ft in the air and the resolution would not be very good. That said, nearly everywhere legal altitude limits are 400ft.
I am trying to determine if it would be feasible to use drone and the DroneDeploy software to monitor the health of my crops.
But my the thing that is bothering me is that all the software that I am looking at is designed and optimized for square fields. That flying an round field is very inefficient.
Lets look at the above example:
-it take 30 min to map one Field and lest say 10 min setup.
-I have almost 100 Fields.
-This it would take about 66 Hours of flying to map all the fields.
-Let’s say I will be able to fly 6 hours a day and work 5 days a week and every day the weather allows me to fly.
-That would mean to image all my fields it would take 2 weeks of only flying the drone to map all my fields this would not be worth it to me since I can’t dedicate that much time to flying the drone.
The other option is to use an fixed wing drone and fly high enough over the flied to only need to take one picture per field to decrease the amount of distance and time the drone has to fly.
It would also help with the uploading of data to the cloud since I am on a farm and data is slow and expensive.
Thanks for the added detail, very helpful. I still think we are speaking different languages. Now you say two pictures per plot? Also, when you say “monitor the health of my crops” that means something pretty specific here that you may not be implying.
To mappers this means gathering imagery on the crops at a much lower altitude so that we can create very high resolution maps of the entire property. Even down to the point as to tell what each kind of crop is and whether or not it is bearing fruit or might be struggling. I am still struggling to understanding if you are trying to fly high enough to capture 1 field in 1 picture from above, 1 field in 2 pictures or if you are trying to take a standard oblique aerial?
On the scale of your fields, Drone Deploy can probably do a good (efficient) job of covering the area. You could also program the flight in Litchi. But your dilemma is not really with flight planning software. It is with the physics involved in what it takes to get imagery that will also inform your decisions.
Setting aside any legality in your area, using a consumer drone, you are going to be limited by the height at which the GSD (resolution of ground detail) is sufficient to give you the answers you are looking for. You can fly higher, take less photos in less time, but the detail will likely be insufficient to tell you anything.
The time you are estimating that it will take to fly all of your fields is probably realistic using 400’ as the baseline.
You could also fly a fixed wing or VTOL hybrid fixed wing aircraft carrying a full frame Mirrorless camera and fly higher with sufficient resolution, but that setup is going to cost more up front. Depending on the camera you use would determine how high you could fly (= less photos / less time) and get sufficient resolution.
Thanks for the information there is my latest flight plan I am considering. This flight plan is 125km in distance and 213 meter in altitude it passes each land a minimum of 3 time and takes on average 16 pictures per field with an 20% image overlap. This all gives me 20cm/px ground res.
I am not to worried about the res because I only want to see issues on the field’s that are over half a meter anything smaller is of no concern percentage wise.
Have you done any testing with your ideas? 20% overlap of homogeneous subject is probably not going to work. Unless you already have tried it?
I am not an expert on fixed wing capture, but I would think that an ever tightening turn and bank angle may be an issue. Also, most processing engines might struggle with basically every image being oriented differently than it’s neighbor. Again, if you’ve already tested this and it is working then great.
Drone Deploy is not going to let you plan a mission like that. Litchi will. You can still process with Drone Deploy if that is your goal.
Non of this is tested only at the research stage, I do agree that the bank in the spiral method would be an issue unless you use a gimbal to keep the camera facing directly down but that’s neither here nor there.
I am not going to be using the spiral method at first I might experiment with it later as for the normal mapping method what is the minimum overlap that is recommended?
It’s important that I try and calculate the amount of distance that would need to be flown in order to determine the size and type of drone that I need to map all the fields in a timely fashion.
For homogeneous subjects, most processors agree on at least 80% front / 70% side overlap.
For the scale of area that you want to monitor, I would bet that a fixed wing setup is what you want. If it were me, I would plan to cover a group of circles at a time. The time that the aircraft is turning around is somewhat wasteful. In fact most processors don’t want images taken in that phase of flight.
Again, if I were in your shoes, just starting out and with what you want to do, I would bring in a contractor already doing similar work in your country and hire them for either a demo or a consultation. You can save yourself a lot of wasted time and money in my opinion.
I think the spiral with a fixed wing is a great way to go! Good idea. DroneDeploy could do that with a quad and the linear flight app. Litchi the same, but also with a couple of orbits with different radii.
Obviously you’ll have a shape to trace, but you can configure an overlap to help you space everything correctly. This is 160ac @ 385ft with 70/60 overlaps. Flight time is about 18 minutes. I think I got some of the lines a little closer than what the 60% overlaps should have been so it should fit in one Phantom 4 battery.