I keep trying to process images taken with smartphones and while it works with ND iPhone and a Xiaomi, I’m trying to process some images of a wooden house walking around the house and I get this error.
All setting are pretty much default using webodm lightning.
File “_qhull.pyx”, line 1841, in scipy.spatial._qhull.Delaunay.init
File “_qhull.pyx”, line 357, in scipy.spatial._qhull._Qhull.init
scipy.spatial._qhull.QhullError: QH6022 qhull input error: 1’th dimension’s new bounds [-0.5, 0.5] too wide for
existing bounds [0.48, 0.48]
I have understood that it’s gps data problem… But I’d like to keep the gps data on the processing and the results…
So, I’ve been looking for another solution
I’ve found a solution… My smartphone isn’t the best and there should be a problem with the GPS tags…
I downloaded Open Camera from Play Store and with it, enabling all the positioning setting the results are pretty well… A little errors with inclination… but nothing too bad.
I didn’t know this option, thanks.
What’s your experience with it?
Isn’t smartphone camera giving a fix for every pic it takes?
Maybe would be good to have a specific photogrametry camera app that keeps gps turned on from the begining and don’t turn it off so we can have better GPS consistency.
This is what ChatGPT tells me about it this settings:
The “force full GPS duty cycle” option in Developer Settings is a feature available on some Android devices that allows developers to test location-based apps and services.
Normally, when an app requests a user’s location, the device’s GPS sensor is turned on briefly to obtain a fix and then turned off again to conserve battery life. This is known as the GPS duty cycle.
Enabling the “force full GPS duty cycle” option in Developer Settings overrides this behavior and keeps the GPS sensor turned on continuously, even when not in use by any apps. This can provide more accurate and consistent location data, but it can also drain the device’s battery more quickly.
In summary, enabling this option is useful for testing purposes, but it should be used with caution in everyday use to avoid excessive battery drain.
Not necessarily. They can use cached or stale data, or data from other location providers using the location API, so accuracy is not guaranteed. They can also just not add location metadata if they can’t get any
ChatGPT should not be used for doing factual research. Large language models are subject to hallucination and cannot be trusted as sources, unfortunately.
Handy when using GPS averaging apps, but in my experience, not much use when taking photos, as the positions in EXIF still only have 30 metre precision, which is of no use most of the time for small area subjects.
