If there’s one thing that never seems to suffer from supply chain problems, it’s litter. It’s everywhere, easy to spot and — you’d think — pick up. Sadly, most of us seem to treat litter as somebody else’s problem, but with something like this machine vision litter mapper, you can at least be part of the solution.
For the civic-minded [Nathaniel Felleke], the litter problem in his native San Diego was getting to be too much. He reasoned that a map of where the trash is located could help municipal crews with cleanup, so he set about building a system to search for trash automatically. Using Edge Impulse and a collection of roadside images captured from a variety of sources, he built a model for recognizing trash. To find the garbage, a webcam with a car window mount captures images while driving, and a Raspberry Pi 4 runs the model and looks for garbage. When roadside litter is found, the Pi uses a Blues Wireless Notecard to send the GPS location of the rubbish to a cloud database via its cellular modem.
Cruising around the streets of San Diego, [Nathaniel]’s system builds up a database of garbage hotspots. From there, it’s pretty straightforward to pull the data and overlay it on Google Maps to create a heatmap of where the garbage lies. The video below shows his system in action.
Yes, driving around a personal vehicle specifically to spot litter is just adding more waste to the mix, but you’d imagine putting something like this on municipal vehicles that are already driving around cities anyway. Either way, we picked up some neat tips, especially those wireless IoT cards. We’ve seen them used before, but [Nathaniel]’s project gives us a path forward on some ideas we’ve had kicking around for a while.
Continue reading “Machine Learning Does Its Civic Duty By Spotting Roadside Litter”
Over in the Swedish city of Södertälje, about 30 km southwest of Stockholm, a pilot program is being explored which will enlist crows to clean up discarded cigarette butts. Butts account for over 60% of litter in Sweden, and the per-butt cleanup cost falls between 0.8 and 2 Swedish kronor each. The company behind the project, Corvid Cleaning, estimates the cost will be around 0.2 kronor. If the birds picked up all the butts, that would be a substantial savings, but in reality, the current manual cleaning will still be needed. Total savings to the city will depend on the ratio of bird-collected vs. people-collected butts. Of course, if people would throw their butts in ashcans or carry pocket ashtrays like those popular in Japan, this would be a non-starter.
Crows were selected because they are considered one of the most intelligent bird — they’re easy to teach, and they communicate with each other. All crows participating in the project are volunteers, and are paid by the butt with a morsel of food dispensed from a machine. We’re reminded of B. F. Skinner’s pigeon-guided missile projects from the 40s and 50s, although cleaning up litter for food should result in a happier outcome for all parties concerned.
This kind of project has been tried before, for example, in a French park back in 2018. And we covered a 2020 project by [Hans] who was training magpies to do similar duty. Are you aware of any of these projects that went past the pilot phase and are in operation? Let us know in the comments below.
Positive reinforcement is the process of getting someone to understand their actions result in a reward. Children get a sweet treat when they pick up all their toys and older ones might get some cash for mowing the lawn. From the perspective of the treat-giver, this is like turning treats into work. A Dutch startup wants to teach the crow population to pick up cigarette butts in exchange for bird treats.
The whole Corvidae family of birds is highly intelligent so it shouldn’t be a problem training them that they will get a reward for depositing something the Hominidae family regularly throw on the street where the birds live. This idea is in turn an evolution of the open-source Crow Box.
For some, leveraging the intelligence of animals is more appealing than programming drones which could do the same thing. A vision system mixed with a drone and a manipulator could fulfull the same function but animals are self-repairing and autonomous without our code. The irony of this project is that, although it’s probably fairly easy to train crows to recognize cigarette butts, the implementation hinges on having a vision system that can recognize the butts in order to properly train the crows in the first place.
If we had the time to train crows, it would definitely be to poop on cars that don’t signal for turns. Maybe some of these winged devices can be programmed to recognize lapses in traffic laws in exchange for some electrons.
Thank you, [jo_elektro], for the tip.