Greyparrot uses computer vision to improve waste management
Meet Greyparrot, a London-based startup that wants to improve waste management. The company uses computer vision to make sorting more efficient at different stages of the waste chain. And Greyparrot has been selected as a wildcard for the Startup Battlefield at TechCrunch Disrupt SF.
The company has been using machine learning with images of different types of waste to train a model that detects glass, paper, cardboard, newspapers, cans and different types of plastics (black trays, PET, HDPE).
Greyparrot can then use a simple camera combined with a computer to sort waste in a fraction of a second.
There are many different use cases for this kind of technology, but it seems particularly promising in sorting facilities. Those facilities already use a ton of machines to separate small and big objects, metal from plastics, etc. But many of them still rely on humans at the end of the process to pick up the last remaining false positive objects.
While it’s never possible to sort everything with a 100% accuracy, you want to get as close as possible to 100%. Sorting facilities create huge cubes of PET plastics and send them to countries on the other side of the world
This post was originally published on this site