Waste-Management

Welcome to Waste-Management! This project leverages machine learning, deep learning, artificial intelligence models, and blockchain to address various aspects of waste management, including waste prediction, classification, disposal techniques, addressing various FAQs, and promoting transparency in waste collection and motivating societies to reduce carbon footprints.

Section	Description
About the Project	Overview of the project's goals, technologies, and approach
Preview of the Website	Visual preview of the website interface
Features	Details of waste management features and prediction models
Smart Contracts	Overview of smart contracts used for waste management and carbon footprint tracking
Overview of CNN Based Models	CNN models used for waste and bag classification
Overview of NLP & SVC Models	NLP and SVC models used for FAQs on waste management
Overview of Sensor Readings Models	Sensor-based models for predicting anomalies and waste status
Overview of ML Models	Machine learning models for waste classification and prediction
Circuit Diagram	Circuit Diagram of our Hardware
Hardware Assembly	Actual Hardware Assembly of the Project
Classification Models	Predictions and Testing of Classification Models
Project Structure	Folder and file structure of the project
Getting Started	Steps to clone and run the project on your local machine
Demo Video	Video demonstration of the project
Contributing	Guidelines for contributing to the project
Contact	Contact information for project maintainers

About the Project

Waste-Management uses sensors and custom datasets to develop models for predicting and classifying different types of waste, managing waste disposal techniques, and integrating voice control features. The backend is written in Flask, and the frontend is created using Next.js. It also uses Solidity to write smart contracts.

Preview of the Website

Features

1. Leak Status Prediction

   • Model: Random Forest
   • Dataset: Custom dataset using MQ-2, MQ-135, and MQ-7 gas sensors.
   • Functionality: Predicts gas leaks if gas value is above 60.

2. High Temperature Alert Prediction

   • Model: Random Forest
   • Dataset: Custom dataset using DHT11 sensor.
   • Functionality: Predicts high temperature if the temperature is above 50°C.

3. Waste Type Classification Based on Weight and Volume

   • Model: Random Forest
   • Dataset: Custom dataset with weight and volume of waste.
   • Functionality: Classifies waste into paper, plastic, glass, metal, or organic.

4. Waste Disposal Technique Recommendation

   • Model: Random Forest
   • Dataset: Custom dataset using moisture sensor readings.
   • Functionality: Recommends disposal technique (0-25: Recycling, 26-60: Composting, 61+: Landfill).

5. Waste Type Classification Based on Moisture Readings

   • Model: Random Forest
   • Dataset: Custom dataset with moisture readings.
   • Functionality: Classifies waste as dry (0-9), mixed (10-30), or wet (30+).

6. Waste Generation Prediction

   • Model: Random Forest
   • Dataset: Custom dataset based on data collected from society waste generation in March.
   • Functionality: Predicts average waste generation, with higher predictions for weekends.

7. Overflow Prediction

   • Model: Random Forest
   • Dataset: Custom dataset using values from an ultrasonic sensor.
   • Functionality: Predicts overflow if value is above 22.

8. Recyclable and Non-Recyclable Waste Classification

   • Model: CNN
   • Dataset: Kaggle dataset
   • Functionality: Classifies waste items in images as recyclable or non-recyclable.

9. Waste Classification

   • Model: CNN
   • Dataset: Kaggle dataset
   • Functionality: Classifies waste items in images into categories like battery, biological, glass, cardboard, clothes, etc.

10. Bag Type Classification

    • Model: CNN
    • Dataset: Kaggle dataset
    • Functionality: Classifies bags in images into garbage bag, paper bag, or plastic bag.

11. Frequently Asked Questions (FAQs WasteBot)

    • Model: NLP with SVM
    • Dataset: Custom dataset with over 700 Q&A on waste management.
    • Functionality: Answers questions based on waste management.

Data Collection From Sensors

Smart Contracts

Overview of CNN Based Models For Waste & Bags Classification

Overview of NLP & SVC Based Models

Overview of Sensor Readings Based Prediction Models

Overview of ML Models for Waste Management Classification and Prediction

Circuit Diagram of Hardware

Hardware Part of Project

Predictions and Testing of Classification Models

Project Structure

• 📁 backend/
  • 📁 Dataset/
  • 📁 models/
  • 📄 app.py
• 📁 pages/
  • 📁 api/
• 📁 styles/

Getting Started

To get started with Waste-Management, follow these steps:

1. Clone the repository:

   git clone https://github.com/JainSneha6/Waste-Management.git
   

2. Navigate to the project directory:

   cd Waste-Management
   

3. Install backend dependencies:

   cd backend
   python3 -m venv venv
   # On Windows
   venv\Scripts\activate
   # On macOS and Linux
   source venv/bin/activate
   pip install flask, flask-cors, sklearn, tensorflow, cv2, pil, joblib, matplotlib, pandas, numpy
   

4. Run the backend server (runs on port 5000 by default):

   python app.py
   

5. Install frontend dependencies:

   cd ../frontend
   npm install
   

6. Run the frontend server (runs on port 3000 by default):

   npm run dev
   

7. Open your web browser and navigate to http://localhost:3000:

Demo Video

Waste.Management.mp4

Contributing

Contributions to this project are welcome! If you have suggestions for improvements or would like to contribute new features or analyses, feel free to submit a pull request

Contact

For any questions or feedback, feel free to reach out:

• Sneha Jain - GitHub | LinkedIn
• Siddhartha Chakrabarty - GitHub | LinkedIn
• Project Repository