This project automates the task of waste segregation in public dustbins by classifying waste as either recyclable or non-recyclable. We use a Raspberry Pi along with a servo motor, IR, Ultra-Sonic Sensor, Pi Cam and a server(laptop) to accomplish the project.
The project has been implemented as an IoT system which captures images through the Pi and sends it to a server for processing. On the server, we use CNNs for the Machine learning/Image processing algorithms, which are used to identify type of waste that was dumped. The server communicates back to the Pi and the Pi actuates accordingly by turning the servo motor and thus dumping the garbage to either the recyclable or non-recyclable side. We use an ultrasonic sensor to detect the level of garbage accumalated in the bin.
If the bin has been filled, we use Twilio's API to send out a message, informing that the bin is full and has to be replaced.
- Files to be placed in Raspberry Pi are in
PI/ - Files to be placed in the Server are in
SERVER/ - Files for Web Server (Frontend) are placed in
CLIENT/ - Directory structures are mentioned within
location.mdfor all Pi, Server and Web Server.
- Tensorflow
- Keras
- SSH along with key setup between Pi and Server
- Python3
- Twilio credentials and libraries
- Setup directories according to each
location.md - Run
python3 ir.pyfrom the home directory in the pi. Machine Learning Model is in Server along with trained weights.
- The model when trained for 200 epochs gave us 89% accuracy.