Garbage Classifier

This project constitutes for garbage detection and classifier within a area using
microsoft custom vision, Azure functions, Raspberry pi and a Camera.

Use of this project lies in the urban places where people throw trash and
garbage on street regardless of having a dustbin nearby.
This project can detect garbages and turn the sound ON whenever it
detects some recyclable or non recyclable stuff.

Login to Azure portal and create a new Custom Vision Instance.
use the dataset in this repo to train the model.

Login to cloudmqtt.com or adafruit.io
to setup a free instance of Mqtt broker.

In production

Fill up and save .env file in func folder
Deploy func as a function app in azure portal
execute in bash terminal export a=0 (on Raspberry Pi)
execute in bash terminal . ./schedule.sh (on Raspberry Pi)
execute python clientMqtt.py (on Raspberry Pi)

Testing Locally

Setup

Install Azure functions core
Fill up and save .env file
Enter details of blob storage account ( function storage account ) in pushblob.py

Run

execute in bash terminal export a=0
execute in bash terminal . ./schedule.sh
go in func folder and execute func host start

Screenshots

File an issue regarding this for anything ✨