Agisoft vs Pix4D in multispectral orthomosaic

[lroque]

hi
The multispectral orthomosaics of the same flight were obtained with Pix4D and Agisoft PhotoScan. On the pix4d index map NDVI the predominant value is about 0.7, while in PS it is about 0.3. Why do NDVI values calculated with both sofwares differ so much?

[Alexey Pasumansky]

Hello lroque,

Currently PhotoScan is not taking into account radiometric information from meta data or from the reflectance panel, so it is the most probable source of the issue. We are currently working on the radiometric calibration feature. If you wish, you can share with support@agisoft.com the sample dataset and processed NDVI map to compare with, so that we could check the planned functionality.

[ken5545]

Alexey Pasumansky,

I am waiting impatiently for this feature. I use RedEdge MicaSense, some sample date you can find here - https://www.micasense.com/red-edge-sample-data

I would like to select what I want to calibrate e.g. only dark current and exposure without BRDF or vignetting because it can be done in other software. Except that, the calibration information must be included in the final report.

Take a look at  'Camera Requirements for Precision Agriculture' from Pix4D and check their XMP tags implementation - https://support.pix4d.com/hc/en-us/articles/206434245#gsc.tab=0