-
Notifications
You must be signed in to change notification settings - Fork 12
meta data is overwirtten with constant data #6
Comments
It seems like someone had at least partial success in reverse engineering the important meta data: https://exiftool.org/forum/index.php?topic=11401.0 |
Hey @najtin Yes absolutely, it would be better if we can extract the formula params from the metadata itself for the DJI encoding too. Would you be interested in contributing the same? |
Yes i will take a look into that. Though i have no experience in reverse engineering and it would be a side project for me. :D |
Haha well, all the best, hope you figure it out! 🙂 |
Hi, Regarding the same issue, DJI released SDK that allows extracting the information from images taken by H20T and XTS but does not yet support M2EA. According to this thread, one of their developers says they will release a version that supports M2EA in the second half of 2021. The formula that DJI used to calculate the temperatures in deg C is different than the one that you should use for FLIR data. |
Thanks for sharing @omritz . Will look into integrating that! |
Hey @omritz - the DJI SDK has been integrated by @agarwalkunal12 for fetching the raw sensor values 😃 |
Can you test it out, and share how it compares to using the DJI Thermal Analysis tool? |
Thanks for the update. When I test it with the new M2EA support and adjust the same parameters, the results are not the same as the DJI thermal analysis tool. |
@agarwalkunal12 - any thoughts on this? Could this be because of the assumptions made for the missing metadata? |
Perhaps this could be because DJI's SDK is not called when M2EA images are processed by |
Hi, i just stumbled upon this project so i am not sure if i understand correctly.
To me this seems wrong:
https://github.com/detecttechnologies/thermal_base/blob/main/thermal_base/thermal_base.py#L110-L125
I am going to try to built it myself and report back.
The text was updated successfully, but these errors were encountered: