3 ms·

I wonder why they're not using thermal cameras. Snakes are ectothermic, but they still show up: https://www.youtube.com/watch?v=EOL0RocIUI0 https://www.youtube.

I wonder why they're not using thermal cameras. Snakes are ectothermic, but they still show up: https://www.youtube.com/watch?v=EOL0RocIUI0 https://www.youtube.com/watch?v=EOL0RocIUI0

burcs 2y ago
Agreed – it feels like sending drones out the levee access roads they mention with thermal cameras would be their best bet.
- skeeter2020 2y ago
  HN: What problem CAN'T be solved with more technology?
atourgates 2y ago
I was thinking about technological solutions. Maybe some combination of cameras (thermal, traditional, maybe even Lidar) combining data and processing it to detect at least likely pythons. It wouldn't have to be perfect, but I imagine such a solution could be more effective than even the most experienced and talented human python spotter.
- MrBuddyCasino 2y ago
  Yes a thermal cam on a drone plus maybe a bit of AI could save a lot of time.
  - skeeter2020 2y ago
    If you're paying people < minimum wage on a per piece basis, is time the dimension where you should focus?
munificent 2y ago
Thermal cameras have a fairly narrow FOV. When you're moving at (albeit slow) driving speeds and trying to see everything around you, trying to look through a thermal camera and scan around would probably be too slow and end up costing you more time. Also consider that there are plenty of other native snakes the same size as a small python that will show up as a false positive on a thermal camera.
- lq9AJ8yrfs 2y ago
  Lenses are a thing in the thermal spectra, but you have a point that they are very expensive if your desired objective doesn't coincide with someone else's bulk use case. They are also typically hard to care for and some of them are toxic. For magnification at least, doesnt help with hunting pythons probably, you can use the focusing lenses for laser cutters, which typically work in thermal wavelengths, to turn a handheld thermal camera into a microscope. Or is it a "milliscope" i guess owing to the limit imposed by the coarser wavelegths. I wonder how much optical solutions could be solved with the "wrong" lens paired with a well trained transformer or similar modern AI. If google or apple can deblur your family photos is it unrealistic to think you could deblur a thermal image acquired using found lenses? Crystallography algorithms might even work out of the box? Or be adaptable hopefully?

I wonder why they're not using thermal cameras. Snakes are ectothermic, but they still show up: https://www.youtube.com/watch?v=EOL0RocIUI0 https://www.youtube.com/watch?v=EOL0RocIUI0

burcs 2y ago

Agreed – it feels like sending drones out the levee access roads they mention with thermal cameras would be their best bet.

skeeter2020 2y ago
HN: What problem CAN'T be solved with more technology?

atourgates 2y ago

I was thinking about technological solutions. Maybe some combination of cameras (thermal, traditional, maybe even Lidar) combining data and processing it to detect at least likely pythons. It wouldn't have to be perfect, but I imagine such a solution could be more effective than even the most experienced and talented human python spotter.

MrBuddyCasino 2y ago
Yes a thermal cam on a drone plus maybe a bit of AI could save a lot of time.
- skeeter2020 2y ago
  If you're paying people < minimum wage on a per piece basis, is time the dimension where you should focus?

munificent 2y ago

Thermal cameras have a fairly narrow FOV. When you're moving at (albeit slow) driving speeds and trying to see everything around you, trying to look through a thermal camera and scan around would probably be too slow and end up costing you more time. Also consider that there are plenty of other native snakes the same size as a small python that will show up as a false positive on a thermal camera.

lq9AJ8yrfs 2y ago
Lenses are a thing in the thermal spectra, but you have a point that they are very expensive if your desired objective doesn't coincide with someone else's bulk use case. They are also typically hard to care for and some of them are toxic. For magnification at least, doesnt help with hunting pythons probably, you can use the focusing lenses for laser cutters, which typically work in thermal wavelengths, to turn a handheld thermal camera into a microscope. Or is it a "milliscope" i guess owing to the limit imposed by the coarser wavelegths. I wonder how much optical solutions could be solved with the "wrong" lens paired with a well trained transformer or similar modern AI. If google or apple can deblur your family photos is it unrealistic to think you could deblur a thermal image acquired using found lenses? Crystallography algorithms might even work out of the box? Or be adaptable hopefully?