Electronics > Projects, Designs, and Technical Stuff
Radar to detect mosquitoes?
phil from seattle:
--- Quote from: RoGeorge on July 30, 2020, 11:10:21 am ---Meanwhile, 10 years ago
:D
--- End quote ---
Yeah, that was Nathan Myrvold's "SDI for mosquitos". Complete blather. Especially since the intent was to protect people in the malarial zones - people that can barely afford running water. This was from the same outfit that wanted to solve global warming by lofting some sort of dust into near-space and reduce hurricane season with evaporative coolers in the Gulf of Mexico. All thought, no action and zero consideration for unintended consequences. Though, they did file a few patent claims.
And while we are at it, the company should have been named Intellectual Vultures, not Intellectual Ventures. But that is the subject of a different rant.
janoc:
--- Quote from: Domagoj T on July 30, 2020, 11:11:12 am ---
--- Quote from: Berni on July 30, 2020, 10:16:48 am ---As others said you need the wavelength to be about as small as what you are trying to see. So about 100GHz gives you 3mm wavelength and an antenna that is about half that size.
Good luck building that.
--- End quote ---
Just go a few orders of magnitude faster and get an off the shelf solution - a camera.
edit:
30 seconds too late.
--- End quote ---
Given the size of a mosquito and the speed with which it moves, that would have to be one heck of a camera to be able to pick them up, especially at a distance of a few meters.
SiliconWizard:
--- Quote from: atmfjstc on July 30, 2020, 02:45:33 pm ---Seems to me more practical to detect them by triangulating their "regular" (i.e. non-ultrasonic) sound emissions.
--- End quote ---
Yep, that's likely to be much easier to implement. Whereas it could be relatively straightforward (with an array of sufficiently sensitive microphones) with just ONE mosquito, being able to track down several mosquitoes in the same area this way would get rather hairy. I guess you could resort to some kind of ICA, but if you add various ambient noises on top of that, not sure how accurate that could be. Now the OP kind of talked about one mosquito, so if there is only one in the room, that could be doable.
Forget about video IMO. Not only do they fly too fast, but you'd need an array of cameras to cover the whole space, with proper focusing so that you could at least identify the location in real time (for very tiny beasts, you'd need a rather good focus). Yeah.
Short of doing some research about how mosquitoes fly, I'm not sure I see the point though. What are you going to gain knowing exactly where a given mosquito is in a room in real time?
magic:
--- Quote from: phil from seattle on July 30, 2020, 03:44:46 pm ---Yeah, that was Nathan Myrvold's "SDI for mosquitos". Complete blather. Especially since the intent was to protect people in the malarial zones - people that can barely afford running water. This was from the same outfit that wanted to solve global warming by lofting some sort of dust into near-space and reduce hurricane season with evaporative coolers in the Gulf of Mexico. All thought, no action and zero consideration for unintended consequences. Though, they did file a few patent claims.
And while we are at it, the company should have been named Intellectual Vultures, not Intellectual Ventures. But that is the subject of a different rant.
--- End quote ---
The malaria thing was to gain press coverage and it worked because you are talking about it.
The patent thing was to earn money if somebody tries to actually build it and sell in the West.
Never take anything coming from America at face value :P
duak:
Since doppler radar detects masses of things that are individually smaller than the wavelength as well as detecting things that are moving, I googled "doppler lidar insect detection" and got quite a few hits. Here's one: https://cosmosmagazine.com/uncategorized/mosquitoes-are-really-on-the-radar/ They were able to detect and count insects and classify them based on their wing beat rate. I'd expect the wavelength of illuminating source to be somewhere in the range of near IR (> 1000 nm) to near UV (>350 nm), basically the visible light spectrum plus a bit.
I see that most of the papers are from the past ten years or so and are of academic interest, so I would expect development of the techniques are still being worked out. I'd expect that as soon as someone with money needs to track individuals, someone will work out the optics, sensors and the DSP to do it.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version