loading...

AI Waste Segregation System

dhananjayanpn profile image Dhananjayan P N ・1 min read

My Final Project

An AI Waste Segregation System that uses Machine Learning and Object Detection to detect the type of waste and segregate it into Biodegradable and Non-Biodegradable depending on the results of the Object Detection. It is a stand-alone system that can be used in a multitude of locations.

Link to Code

GitHub logo Dhananjayan-PN / AutoDust

AI Waste Segregation System to clean the world one AutoDust at a time

AutoDust

An AI Waste Segregation System that uses Machine Learning and Object Detection to detect the type of waste and segregate it into Biodegradable and Non-Biodegradable depending on the results of the Object Detection. It is a stand-alone system which can be used in a multitude of locations.




How we built it

The system was built entirely in Python. For the Object Detection, we used YOLOv3 as our trained model to accurately detect the type of waste. We used a RaspberryPi to control the motors depending on the command from the server. Yes, we divided the workload into 2, the server (cloud) and the client (RaspberryPi), which made it much faster and efficient.

Additional Thoughts

We learned a lot of new things in this project including how to use the Cloud (thanks AWS) to your benefit. We also learned how we could efficiently divide the workload for better performance. All-in-all great experience!

Posted on May 20 by:

dhananjayanpn profile

Dhananjayan P N

@dhananjayanpn

Programmer, App Developer, AI Enthusiast and Music Lover

Discussion

markdown guide