What I want to do
I want to understand how others which do NDVI with one or two cameras decide which exposure time to use.
My attempt and results
Setup Two cameras, one normal and the otehr IR lock filter removed and a 720NM filter in front of the lens. Result images are converted to jpeg to save disk sapce. Both chdk raw output Vis: shutter 1/200 iso 200 NIT: shutter 1/200 and 1/400 iso 200
VIS for NIR 1/200
VIS for NIR 1/400
NIR 1/200
NIR 1/400
NDVI 1/200
NDVI 1/400
Questions and next steps
As you can see, the 1/200 NDVI picture has better indexes. My question is: How do we determine the correct exposure value for both cameras. Maybe 1/100 was better. Does anyone have an idea.
One solution which just pop up after my post: Make a picture both visible and NIR and let the camera decide what are the best iso and shuter settings for current lighting conditions. With chdk lock these settings, ISO and shutter speed on both cameras. With these "calibration" every picture made on diffrent days should yield almost the same NVI result. And Hope the shutter speed is fast enough for the movement on my drone or pictures will be blured. Comming days only rain .
Why I'm interested
If I want to use this NDVI pictures with farmers they cannot differ to much between diffrent days.
1 Comments
I often use platforms that are moving so I use a fast shutter speed to reduce blur. I often shoot somewhere around 1/1000th of a second. You should be able to compensate for different shutter speeds so the resulting NDVI images are similar even if different shutter speeds are used. If you have a reference target in images with different shutter speeds you could calculate the pixel value difference and then adjust the other image by applying an offset. It's possible that correct would require more than a simple offset but that's a good place to start.
Reply to this comment...
Log in to comment
Login to comment.