VIs from uncalibrated NIR and VIS images


I've been working with UAV imagery (NIR and VIS) for monitoring sugarcane's post-harvest behavior in Guatemala. The images are uncalibrated and I'm using raw DN numbers for extracting NDVI and other VIs. The main purpose of the flights and analysis is count (or estimate, at least) the number of plants per row in the field. At this post-harvest stage, grass-type weeds and broad-leaf weeds are growing too, which makes the use of uncalibrated images a little bit tricky. 

Anyone else working with this kind of stuff and having some experience dealing with crops and weeds during early stages of the plant?


You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –


  • I am working on detecting and differentiating weeds between crops in the early season here in North Ontario. Originally high spatial res. hyperspectral data would be the way to go... as you have seen weeds and crops in early phenological stages are spectrally very similar. collecting endmembers with a field spectrometer and performing a supervised classification to determine weed locations vs crop is a good way around this.

    In your case I'm interested to here what you find- would you mind updating this forum page when you get more progress?

    I've also seen OBIA (object based image analysis) with ecognition software to differentiate weeds from crops ....however this doesn't work for the weeds in the seedling rows.
    This domain may be for sale!
  • Jose,

    You are in good hands with Phil, he is a great source of information.

    That palm count example reminds me of that thesis about palm segregation from satellite pictures :-)

    • You can "count" on us......     ;/

    • Good morning Phil:

      Thanks for the reply, I will try to get in touch with you tonight via Skype.

This reply was deleted.