Hello Fellow Yuneec Pilot!
Join our free Yuneec community and remove this annoying banner!
Sign up

H520 Mapping - Sweetwater Section N

Joined
Nov 28, 2019
Messages
524
Reaction score
308
Age
49
Location
Austin, Texas
Just a few images of a project that was recently flown using DataPilot the H520 with and Emlid PPK solution. Stakeout errors were +/- 3cm and we used 6 GCP's on the 90 acres for positioning and the GSD is 0.8in/px. I don't know if this is the right place to post a project share on this forum so a moderator can move it if needed.

Base Orthomosaics and CAD Overlays
1575047068994.png 1575047288903.png 1575048521025.png

Stockpile Volumetrics for Materials Tracking and Accounting
1575047441367.png

DSM Point Cloud (We create our own DTM's with different software in-house for further analysis)
1575047658235.png 1575047871919.png

3D Textured Mesh
1575047961180.png 1575048328238.png
 
It looks good. Thanks for sharing ?

I see something unusual, in such a large area, 6 points is not a few? The undulations in the lines of the highway will not be for lack of GCP's?

Out of curiosity, how many images have been processed and with what software?

I have more questions, but little by little ;)
 
It looks good. Thanks for sharing ?

I see something unusual, in such a large area, 6 points is not a few? The undulations in the lines of the highway will not be for lack of GCP's?

Out of curiosity, how many images have been processed and with what software?

I have more questions, but little by little ;)
Thank you. There were approximately 600 images processed with DroneDeploy. The roadway was a previous project we did and was not part of this survey so it is on the edge of the map and only had about 50% side overlap. This flight was solely nadir images and only 6 GCP's were needed because the images were geotagged using a PPK GNSS workflow. I'll post more about that process on another thread. This mission was flown to capture ground progress and to quantify the stockpiles and amount of curb and gutter that had been installed.
 
With little overlap, it's normal to have those undulations. Waiting for the workflow. Really, in the real projects is when the real problems arise. On paper it's all very nice. Then you have to solve what comes up, and see how others solve problems in different ways, is what really makes you learn and grow. Thank you.
 
Looks good, how many battery swaps did this take and what elevation did you fly?
We just purchased an Emlid system for GCPS, did you use a Reach M module on the drone for PPK corrections? If so was it hard to set up?
 
This was one flight with 75/75 overlaps. With a P4P it does take not much, but a second battery whereas it is on that threshold and the H520 can do it in one. Flight AGL was 225ft. I used the Emlid Reach RS+ to capture the horizontal of the GCP's, but because Reachview only captures the ellipsoid I use MicroSurvey FieldGenius so that I can collect the control points from our surveyor's Topcon localized control and get the correct ground elevations. In instances where I am doing work outside of our company I just need a plat from public record or some points established in QGIS and some type of vertical benchmark to do the same localization for the GCP's. The importance is that even with RTK/PPK I need to align to the survey and design files that came before in order to be effective down the line. This happens with GCP's and anyone that says they are doing this without GCP's is not familiar with what we are doing.
 
  • Like
Reactions: arruntus
Hello, I have an H520G (which uses Yuneec ST10C controlled + iPad + SiteScan iOS app). This is a pain to use for manual camera settings as it won't let you change them while you're up in the air. Thus far I've just flown all missions on auto camera settings.

Today I decided to compare the output orthos of the H520G with those taken from a P4P. I did two exact flights using SiteScan iOS app, back to back, only that for the P4P I used manual settings: shutter priority, ISO 100, and Auto AWB. This was at 300ft altitude and 80% overlap, 80% sidelap.

The H520G pictures looked overexposed and almost soft / no good color. The resulting ortho was pretty terrible, the trees were more like blobs and you could not define the leafs very well. On the other hand, the P4P outputs were amazing. The images were crisp and colorful. The trees came out stitched almost perfectly. Are there any settings/workflow you follow for this? Any help would be extremely appreciated.
 

Members online

No members online now.

Forum statistics

Threads
20,973
Messages
241,794
Members
27,357
Latest member
Bech