Smart Shelf Management System This project aims to automate shelf management using computer vision and IoT technologies. It tracks product quantities, detects objects using machine learning models (YOLOv5), and manages stock levels through an inventory system.
Features Product recognition using YOLOv5 and ESP32-CAM
Real-time inventory tracking
Stock level prediction using time-series models (LSTM/Prophet)
Demand forecasting using machine learning algorithms (Random Forest, XGBoost)
Flask-based backend for data processing and API
Tech Stack Computer Vision: YOLOv5, OpenCV
Machine Learning: TensorFlow, Keras, scikit-learn
Backend: Flask
Database: SQL-based (SQLite or MySQL)
IoT: ESP32-CAM for image capture
Data Prediction: LSTM, Prophet, Random Forest, XGBoost
Getting Started
-
Clone the Repository bash Copy Edit git clone https://github.com/your-username/smart-shelf-management.git cd smart-shelf-management
-
Install Requirements bash Copy Edit pip install -r requirements.txt
-
Start the Flask Server bash Copy Edit cd server python app.py
-
Set Up ESP32-CAM Flash the ESP32-CAM with the appropriate firmware to capture images and send them to the Flask server.
-
Run Object Detection Upload product images or stream live video from ESP32-CAM to detect and update inventory.
Project Structure
smart-shelf-management/ │ ├── model/ # YOLOv5 and object detection code ├── server/ # Flask server and database integration ├── esp32/ # ESP32-CAM firmware code ├── prediction/ # LSTM, Prophet, Random Forest models for prediction └── README.md # Project documentation
Future Scope Integrate barcode scanning for faster product detection
Implement real-time notifications for low stock levels
Add predictive analytics for inventory management
License This project is licensed under the MIT License.