Photogrammetry with gluonpilot

[Tom]

It is implemented in the latest release (or beta?) Replace Survey with ZSurvey to test it!

[kempo]

It is implemented in the latest release (or beta?) Replace Survey with ZSurvey to test it!

NICE!!!

I will test it during the weekend.

http://youtu.be/qK47q8YBKjM

I wonder how many other hidden cookies do you have in the Ground station

[kempo]

Hello all,

Unfortunately , due to bad weather, I did not had time to check the Z-type flightplan but, during the analysis of the the flights a question jumped : how I could correlate the picture made by the camera with the gps position of the camera( gps position from the autopilot) ?

There are some tricks already implemented?

On the link below I have posted some flight analysis using the X5 wing:

http://www.myap.ro/gluonpilot-photomapp ... ript-v2-2/

[Tom]

Nice website!

Yes there is a tool GluonMapper that generates project files for Dronemapper or Pix4D. Do you need it?

[kempo]

Nice website!

Yes there is a tool GluonMapper that generates project files for Dronemapper or Pix4D. Do you need it?

Hi Tom,

If the tool could be shared I think that it could help the community.

Thank you,
Alex

[kempo]

Good evening,

Some real tests using Z-type Survey:

https://www.youtube.com/watch?v=zeGXSAJStKE

From my point of view it it much better than the normal type of Survey, for the tests done above the wind amount was very small, I will retry with worst wind condition.

Now on the TO DO list remains the insertion of the autopilot GPS coordinates to the pictures.

[Tom]

Nice flight! You must be one of the first to try it out in a real flight

[Tom]

Please unzip and start mapperwizard.exe

[kempo]

Tom, thank you for the software !

I have started to play with the autopilot logs ( from board) and I discovered a lot of interesting stuff .

On my first trial I have associated manually to the pictures the gps coordinates (using the time when the picture was taken on the camera and the corresponding lat/long/altitude from the autopilot at the same time).

The results seems good:





Next steps:

1. insert the gps from above to the corresponding pictures - exif;

2. compare the picture gps values from manual mode with the auto mode (software from Tom) to check the accuracy - to be honest I do not know how the software calculates the gps coordinates for each pictures taking into consideration that I have pictures taken at 3 sec delay or even 4 sec delay in the same flight ( on the script the delay is 3 sec).

Function nice to be implemented for photomapping :
- area covered by the flightplan;

Thank you again Tom!!!

[kempo]

Hi all,

* I have made more analysis , this time by comparing in the same .kml file the flight log, pictures tagged auto and pictures tagged manually:



There are 2 types of tagging cases (examples above with black square) :
- for the same picture the gps tagging in manual mode is earlier than the gps tagg added automapically;
- for the same picture the gps coordinates are in the same place.

* Also I wanted to check if the picture is somehow correlated with the "real ground". As reference I have used the Home position ( I was there , small mark on the ground).





If i look at the .kml file to see what is the picture that theoretically catch me on the ground it should be IMG_6015 , but checking the pictures on IMG_6015 i'm not there , but I am on the IMG_6012.... Now the big dilemma starts , why, if the pictures are tagged correctly i'm not in the image ?

I am adding to the "function nice to be implemented" another request :
- the "projection" of a picture to the ground on Google earth ( for a picture, I know the longitude, latitude, altitude, roll, pitch , yaw), something like here :



( the picture is from diydrones forum : http://diydrones.com/profiles/blogs/geo ... 3#comments )

Any observation, advice is welcomed.

Thank you,
Alex