Taking time stamps from image exif to calculate position

Would it be possible to put the Reach RTK on a UAV (totally independent of the UAV) to constantly log position during flight, then later (after flight) get all the time stamps from the images exifs and combine them to get time stamps in the Rinex output file?

Perhaps trough an automated script?

Yes. You can set up Reach to record when a photo is taken by linking to the hot-shoe of the camera. You will need to reconcile that event log with the images because it won’t be recorded in the exif data. You can do that with a script.

https://emlid.com/reach/

Dave,
i know about the hot-shoe combination, I use it with my M600PRO.

But the question goes for UAVs that don’t have the possibility to output the triggering moment, like the DJI Phantom series.

Would it be possible to put the Reach rover on a Phantom, let the Reach constantly log position during flight and then later:
a) get the time stamp from the image exifs from the Phantom and put them in a CSV or alike format
b) make a script that would take this CSV and allign them with coordinates from rover output log
c) do PPK with with data from correctional base stations

@dario.finderle
I think you could make that work if the the clocks between the 2 gps units were synchronized. Simon Allen has built a kit for the Reach that captures when the image is taken and records that in the Reach log.

@dpitman
I’ve read the article but it’s unclear how the capturing is done. I see some arduino electronics but I don’t see how it’s coupled to the Phantom.

Overall, to understand the time synchronisation, why is there an time difference in two GPS receivers that are placed side by side? Correct me if I’m wrong, but after distance to satellites calculations are done, the GPS receiver gets it’s time from the clock send from the satellites. The time when signal reaches both GPS devices is allmost the same. Or not?

Dario, I don’t know if there is in fact a time differential. I guess in theory, you could just find the closest location point from the log to the time stamp in the image exif, mix in some distance from gnss antenna to camera sensor and call it good. I don’t know. Maybe @Simon_Allen will share his thoughts as he has been down this path already. I suspect if it were a viable solution to just compare logs then he wouldn’t have bothered designing his timing circuit

The time that is stored in the EXIF header is the creation date (the time it was stored) not the time it was recorded. Depending on your SD card and the load it is under this may be between 1.5 and 3.5 seconds after the image was recorded. If you use Pix4D caqpture then there is a UTC millisecond time recorded in the app associated with each image capture event, that time suffers from a delay based on the transmission time from the drone, which varies with quality of comms and is unclear which time base is used, the phones or the drones. Phone clocks are close (but not perfect) to UTC depending on when and where they last synched. At 10m/s a 1ms delkay = 1cm
I capture the time sync on the drone directly to the reach using the red lights blink off during capture. The arduino is used to debounce the signal and provide a properly signal for the reach. It can also be used as an external sync to a second IR camera.

This topic was automatically closed 100 days after the last reply. New replies are no longer allowed.