What is the minimum overlap necessary for a survey of something like apple trees? They don't get much taller than 10 feet, and for my clients, they aren't looking for volumetrics, just plant health. I know from VR stitching experience that 20% overlap is the minimum, with 30% being a sweet spot. I'm currently shooting with 60/60, but if I could lower that and still get good stitched results, then my missions would fly much faster, (fewer circuits). Also, the cost of the stitching would be lower as well, due to the reduced number of images, right?
Thanks,
Mark
In previous versions we allowed the Inspire Pro to take images at up to one image per 2 seconds but had way too many users experiencing issues with images being skipped. We kept it that way because the X5 was the only option for certain things. Now that the X5S is out and triggers suitably fast we are now enforcing DJI's specified limits for the X5.
Vous devez vous connecter pour laisser un commentaire.
On a somewhat related note, I've found using another piece of software, that I can fly the Inspire 1 Pro up to 12 m/s and manually fire the shutter every 1.5 seconds and not miss a frame. I realize the DJI SDK limits the Inspire 1 Pro to one frame every 5 seconds, but the potential is there for faster firing, and therefore, faster flying speeds.
Just my $.02
Mark
I agree that being conservative and getting all the frames is the highest priority. It was just interesting when I did my testing that I could fly at 12 m/s and manually push the shutter button on the radio every 1.5 seconds. I even looked into having someone make a mechanical shutter trigger that would mount on the radio to push the button based on an intervalometer. Apparently, it's not too difficult to do that.
If I were to go that route, is there any way that Map Pilot could be set up to fly faster, leaving the camera triggering to an external mechanical finger? That would be a huge step forward in the efficiency of the Inspire 1 Pro, as I could cover the same amount of survey ground in significantly less time.
However, my primary question was about front and side overlap.I realize that lowering the front overlap wouldn't result in a speed increase, but lowering the side overlap would reduce the number of passes necessary for any given survey, thus reducing the time to fly that survey. I could see me doing a front overlap of 50% with a side overlap of 25%, if that would indeed be enough to get a good stitch. As I said, I'm not interested in volumetrics or 3D information from these surveys, just plant health.
Thanks,
Mark
Just a follow-up. What is the lowest side overlap that will yield good stitch results? Is 20/20 too low overall? Would 50 front and 20 side work?
Thanks,
Mark
Based on the stitching results from my first missions, 60/60 is not enough, especially in forest areas. I'll bump up to 75/75 in the future.
