Just got a batch of 4 webcams from the Korean company Huentek; Here's the HU203, and I'll post more as I go:
Here's the original:
But in Infragram Sandbox I seem to be able to pull out a reasonable vegetation differentiation, although this may just be because the vegetation is brightest in this scene:
We really need a more standardized testing methodology for webcams; some kind of test plot with grass, stone, dirt, etc.
Here's the HU106 -- not as good as the HU203:
The auto-focusing HUAF303 is more expensive ($18 in bulk), but seems to work quite well; hopefully we'll be able to test this out more thoroughly at LEAFFEST because it seems that cardboard also has a high NDVI value, but I was able to get decent differentiation: