Creating Orthophoto from MULTISPECTRAL images

Hello to all,

Is it possible to process MULTISPECTRAL images in WebODM? I have tried to create orthophoto from MULTISPECTRAL images but every time i had an error.




I am also trying to create an orthomosaic with multispectral images from a parrot sequoia camera. The goal is to use this in qgis and calculate NDVI index for multiple agricultural experimental plots (this part I can do)
Here are some files I tried to use: multispectral images - Google Drive
I am new to this so I don’t really know which parameters to use to do it. I tried but always had a process exit with code 1 error. And this after 5 hour of computing when pix4fields does it in 10 minutes on the same computer. Maybe I am asking the software to calculate more than what I need?
Could somebody help me with the parameters and/or data to select ?

Thank you for your help


1 Like

Hello everyone!

Same problem: I want to construct orthomosaic with multispectral photos taken with the DJI P4 Multispectral with RTK station. The images generated by each band are .TIFF files and the RGB is jpg. However, when I use the multispectral processing option in WebODM, bugs appear (Code 1 error). Likewise, if fort include support for this type of cams you need a set of data, I can send it to you without problem.



Hello Hector,
Can you send me picture set of one of your DJI multispectral mission?


Hello GDimitroff

Sorry for the delay on my response.

Here two links to datasets

I wait your comments

1 Like

You have to sort out the JPGs and process them in a seperate task. Processing the tiff images (mission 1) with standard multispectral settings works fine.
Unfortunately there is a heavy change in brightness (clouds?) between the pictures.


Hello Hubertus

This drone DJI P4 RTK Multispectral has “an integrated spectral sunlight sensor on top of the drone captures solar irradiance, which maximizes accuracy and consistency of data collection through different times of day” (see here). Have you setted this variable
in the process?

It is worth mentioning that these same data produce correct orthomosaics (without variances in pixel intensity) when processed in the drone manufacturer’s software called DJI Terra.

I wait your comments.


Hello Hector,
the feature of radiometric correction by using the data from a light sensor is still experimental in webodm. Usually this data is stored in a seperate file and not in the exif header of the pictures.
Also this sensor data cannot correct changes in the reflection intensity caused by shadows on the ground.
The paramter for the radiometric calibration is set to “camera” when using the “multispectral” preset and also the global seam levelling is switched off. You can try to run the process with different settings for this parameter.

1 Like

And in every photogrammetry software …
It’s not reliable and even worser than a calibration target.

1 Like

anyone can fixed it pls suggest me about how to fix it