Near InfraRed

ralddischen

Hey,

i need NDVI for my work (in ArcGis and ecognition).Till now, i have always worked with four band digital imagery - but for this work i have three band orthophotos and "three" band near InfraRed orthophotos.

All channels together (near InfraRed Ortho)

[ATTACH=CONFIG]30975[/ATTACH]

just the red channel (InfraRed Ortho)
[ATTACH=CONFIG]30976[/ATTACH]

My question: which channel should i take for the NDVI calculation? (NIR-Red)/(NIR+RED)?

NIR = all three channels together, or just the red one???

Thanks for any help!

rot.jpg

Neues Bild.jpg

Anonymous User

Original User: anlrober

Hey,

i need NDVI for my work (in ArcGis and ecognition).Till now, i have always worked with four band digital imagery - but for this work i have three band orthophotos and "three" band near InfraRed orthophotos.

I assume that the 3-band NIR orthophotos are basically a RGB camera with the NIR cutout filter removed and a NIR filter installed. Digital cameras are generally more sensitive in the 'green' band because human eyesight is more attune to the region in the electromagnetic spectrum. Thus, you want to use band 2 in your NIR imagery (assuming it is set up like a typical RGB camera).  While not specific to orthoimagery, some of my research has used digital camera systems for monitoring vegetation. Do understand that camera-derived NDVI is NOT the same as reflectance derived NDVI. We go into some detail in our paper on why this is the case. Camera-based NDVI is more like the VI wide dynamic range vegetation index (WDRVI).

Sakamoto et al. 2012. An alternative method using digital cameras for continuous monitoring of crop status. Agricultural and Forest Meteorology. 154-155: 113-126.