Welcome, Commercial Drone Pilots!
Join our growing community today!
Sign up

Invalid drone data fix?

roto116

New Member
Joined
Jul 30, 2020
Messages
2
Reaction score
0
Age
29
Has anyone had issues with Pix4D processing due to bad drone data? Sometimes, I realize only when I attempt to process the data that the captured drone data was invalid due to bad brightness or missing GCPs and I end up having to go back to site to refly which causes a lot of delay. Does anyone else face this issue? Any apps that can help? Is there a way to know that the data captured is bad/incomplete right after completing the flight?
 
Sorry for your problems, unfortunately I can't help you but I am sure other members here will give you some input and welcome to the forums - a lot of good info here. So, take a look around and hope you enjoy some of the fine photography and information posted by the members.
 
Has anyone had issues with Pix4D processing due to bad drone data? Sometimes, I realize only when I attempt to process the data that the captured drone data was invalid due to bad brightness or missing GCPs and I end up having to go back to site to refly which causes a lot of delay. Does anyone else face this issue? Any apps that can help? Is there a way to know that the data captured is bad/incomplete right after completing the flight?

Define "bad drone data". What exactly is wrong. That is the only way you are going to get any worthwhile help. The devil is in the details.
 
You mention "bad brightness or missing GCPs". It may also be helpful if you describe what your current operations workflow is. What apps you are using etc.
 
You mention "bad brightness or missing GCPs". It may also be helpful if you describe what your current operations workflow is. What apps you are using etc.

Sure, let me elaborate more on the workflow and the problems that I am facing. On several occasions, I have rented a drone (P4P v2.0, Mavic 2 Pro) to execute mapping projects and I have noticed that the drone camera settings are modified. I use the Pix4Dcapture app to execute missions. Due to poor lighting conditions (cloudy or flying at the end of the day) and the non-ideal camera settings, the images turn out to be too dark. Due to the short flying window and the large aoi that I need to fly, I sometimes forget to set the right camera settings. I don't realize this issue until I am back home and have started processing.

Another issue I typically face is with GCPs. When planning a drone mission, I distribute the GCPs using Google Earth and export them as a KML file that is given to the DGPS vendor on-site. However, there have been situations where some GCPs have been removed / not placed by the vendor. This results in my data processing accuracy to reduce at those spots. This issue in particular is painful because it is near impossible to go through the drone images manually and check if there are GCPs visible or not.

I am interested in knowing if anyone has found a solution to better identify these issues post-flight and take quick corrective action before leaving the field.
 
In my opinion it sounds like you are approaching this the wrong way. I would start by addressing the issues to eliminate them on the front end (in the field) rather than doing damage control on the back-end (processing). Think "junk in = junk out"
Due to the short flying window and the large aoi that I need to fly, I sometimes forget to set the right camera settings.
Create a checklist that includes each of the camera settings and make sure you verify them before flying. We reference a checklist before every takeoff to make sure everything (camera settings, drone settings, etc.) is where it needs to be before you fly.

The same goes for the GCP's. Communicate with whomever is placing the GCP's for you and explain to them that they are necessary to deliver quality data. If they are not placed how you have specified before the mission then you cannot guarantee the results. That should be a pretty easy one to address I would imagine. As a last line of defense, since you know where the GCP's are supposed to be, I suppose you could do a manual flight to each GCP location before you start mapping so that you can verify its presence.
 
  • Like
Reactions: R Martin
In my opinion it sounds like you are approaching this the wrong way. I would start by addressing the issues to eliminate them on the front end (in the field) rather than doing damage control on the back-end (processing). Think "junk in = junk out"

Create a checklist that includes each of the camera settings and make sure you verify them before flying. We reference a checklist before every takeoff to make sure everything (camera settings, drone settings, etc.) is where it needs to be before you fly.

The same goes for the GCP's. Communicate with whomever is placing the GCP's for you and explain to them that they are necessary to deliver quality data. If they are not placed how you have specified before the mission then you cannot guarantee the results. That should be a pretty easy one to address I would imagine. As a last line of defense, since you know where the GCP's are supposed to be, I suppose you could do a manual flight to each GCP location before you start mapping so that you can verify its presence.

Nailed it.
 
Has anyone had issues with Pix4D processing due to bad drone data? Sometimes, I realize only when I attempt to process the data that the captured drone data was invalid due to bad brightness or missing GCPs and I end up having to go back to site to refly which causes a lot of delay. Does anyone else face this issue? Any apps that can help? Is there a way to know that the data captured is bad/incomplete right after completing the flight?

Based upon the additional data you provided, the answer to your question is no. There isn't a magic wand you can wave to undo what you've done. You are stuck with the data you collected. Learn from it.
Suggestions I would make are create a series of checklists that walk you through the process from flight planning, airspace authorization (if required), preflight UAS inspection, mission planning and mission setup, a pre-launch checklist to double check settings, site conditions, weather on-site and anything else you deem necessary (GCP placement).
Next, follow the checklists each time every time. Just when you think you've got it down you'll miss something and that usually means flying the job again (and losing money).
There are a lot of things that go wrong even with meticulous planning. Don't add to your problems by introducing the little mistakes that are entirely avoidable.
 

Members online

No members online now.

Forum statistics

Threads
4,278
Messages
37,606
Members
5,970
Latest member
parker