DIY Noir Insect Camera Trap, solving the MIB paradox, looking for some advice.

Hello fellow Nerds, Geeks, and, well, engineers.

I am developing a camera trap to register diurnal and nocturnal visitors to animal feces.

*This is my challenge right now ->

Lots of small creatures won't be detected by PIR / Infrared Motion Sensors so…. I am counting on computer vision to detect the little bugs however, I wonder…

As, supposedly, I am going to use a MOSFET Module AOD4184 to enable the Raspberry Pi Zero to switch on/off my infrared lights powered by a 12V battery, my question is, how could the Noir camera “see” any movement to then activate the lights, if it can only see anything if the lights are on?

Am I right? Or totally wrong? I call it the Mosfet Infrared Bugs Paradox or MIB Paradox for short.

Any of the many geniuses out there could give me some “light” regarding this topic?

Thank you all so much, bst




Adrien Pajot
@Adrien_Pajot  | He/His
WILDLABS
Fauna & Flora
Hi! I am Adrien, a dedicated French ornithologist and engineer committed to biodiversity conservation. I joined the WILDLABS team as a project manager in October 2023!
WILDLABS Team
WILDLABS Event Speaker
Variety Hour Regular
Inventory Beta Tester
Involvement level 3

Hi Sandro, @eferguson had a project in 2024 to try detecting insects on camera traps. 

Here is the thread: https://wildlabs.net/en/discussion/wildlabs-awards-2024-enhancing-pollinator-conservation-through-deep-neuralnetwork