Phantom 4 multispectral sunlight sensor

[_danny_]

Hi, just wondering if anyone using the phantom 4 multispectral has been able to incorporate the gain settings captured from the sunlight sensor for radiometric calibration? I can't tell if metashape automatically utilises this data or not!

[smhese]

interesting topic.

I am also using a P4M and so far its doesnt seem to be possible to calibrate the P4M data with a Micasense grey reference target. The panel calibration values seem to be totally wrong. I guess that the sensor data isnt fully supported by Agisoft. Any news from Agisoft? Is there a date when he P4M will be supported for reflectance calibration. We are running a 1.6.2 license cluster. Kind regards! Sören Hese (Friedrich-Schiller-University Jena)

[Alexey Pasumansky]

Hello Sören,

Can you please share the minimal dataset from DJI Multispectral for testing and send the download link to support@agisoft.com?

The complete documentation for the meta data for DJI Multi-spectral is missing, however, we have implemented the reflectance calibration procedure for it, according to our knowledge and information we are able to get. Currently the recommended way is to include the radiometric panel to the calibration process anyway, even if you are using Sun Sensor option.

[smhese]

Ok thxs - yes will do. We collected a a full 5 channel tif dataset for 200 ha but I will send a ZIP with 100 images in multi cam system layout at import and the micasense panel tifs. Kind regards! Sören

[droneservicesireland]

Any updates from anyone on calibration panel with P4M?

[Eckemon]

The problem still seems to exist. The results are completely off set when using a MicaSense calibration panel and can't be used scientificly. Are there any updates for a possible solution?

[Alexey Pasumansky]

Hello Eckemon,

Are you comparing the index calculation results (like NDVI)?

[Eckemon]

Hi Alexey,

no, so far I compared the values of the individual bands of multiple orthophotos of the same area. As stated before, the calibrated values seem to be completely wrong and differ highly when comparing them on the same coordinates. Even if the mission was flown on the same day under very similar weather conditions.
I was processing them with the same parameters using the calibration panel and the sun sensor option.

Edit: Same applies when comparing the NDVI (not as bad, though). I sent some example data to support@agisoft.com (see PM). If you need additional data just let me know!

Thanks for your help.

[Alexey Pasumansky]

Hello Eckemon,

The difference in absolute band values is expected, as DJI guidelines provide the equations accurate to coefficient, which goes away during the NVDI calculation. However, the difference of NDVI values seems considerable for the provided dataset, depending on the reflectance calibration options used.

I have reported this issue to our DJI contact, we were working with on vignetting correction for P4 Multispectral. I hope we will be able to track down the possible reasons of such difference, and if it is related to internal Metashape processes, we will fix them

[Eckemon]

Hi Alexey,
thanks for your answer.
Were any changes made in version 1.7x?
I'd appriciate any update regarding this topic!

Kind regards,
Simon 

[Christina K]

Same question as post above.

[droneservicesireland]

Any fixes to this with the recent FW upgrade on P4M ?

[Eumema]

Any chance there is somewhere a workflow document for this? I am struggeling with best practise how to process these Phantom 4 multispec images that I got - they are seasonally taken from the same location. Thanks a lot