3 ms·
This will revolutionize greenhouses and aquaculture. Now you can optimize everything based on plant growth rate. Just need a few million data points and we can
by BatFastard 9y ago
This will revolutionize greenhouses and aquaculture. Now you can optimize everything based on plant growth rate. Just need a few million data points and we can sic the AI on it.
- tw1010 9y agoI don't think you're wrong, but I suspect the reason you're being downvoted is because of the cliche AI-enthusiastic rhetoric.
- BatFastard 9y agoBut it's a great application for AI, or would they prefer ML or DL? Its a dataset with near real time response. As for downvotes, I try to add to the discussion, if people don't like it, oh well, it's only numbers. (unless there is a secret ritual for people who go over 100,000??) [ed] Speeling correction
- RosanaAnaDana 9y agoWell, we do use ML quite a bit in scene classification, but as I said earlier, there is really nothing revolutionary about what was being presented in that article. Its all old science and methods that have been in employ for decades. LiDAR and hyperspec is the way to go, but as is ALWAYS the case, its all hand-waving without good reference data.
- RosanaAnaDana 9y agoWe already do remote sensing in these wavebands and chlorophyll fluorescence spectroscopy has been around a long time. The amount of signal due to chlorophyll fluorescence is very very small. I think if your going to spend the bucks on an advanced sensor to do this kind of work, you just go whole hog and do hyperspectral imaging.