✋ Real-Time Hand Gesture Recognition Using Deep Learning

This project implements real-time hand gesture recognition using a webcam feed. A LeNet-based Convolutional Neural Network (CNN) is trained to classify 9 distinct hand gestures. The application overlays the predicted gesture and its probability on the video stream in real-time.

---

🧠 Model Overview

A classic LeNet architecture (defined in lenet.py) is used for gesture classification. It has been adapted for recognizing hand gestures efficiently.

---

📁 Project Structure

├── lenet.py # LeNet CNN architecture
├── train_network.py # Trains the model and saves it as mask.model
├── test_network.py # Tests the trained model on sample images
├── main.py # Real-time gesture recognition via webcam
├── dataset/ # Training dataset for 9 hand gestures
├── images/ # Sample gesture images for testing
├── mask.model # Saved trained model

🔧 Requirements

• Python 3.x
• OpenCV

  pip install opencv-python
  
  
  

▶️ Running Real-Time Gesture Recognition Make sure your webcam is connected, then run:

python3 main.py

🧪 Training the Model

• To train the CNN from scratch:

    python3 train_network.py
  

• The trained model will be saved as mask.model.

📷 Testing the Model

• To test the model using sample images in the images/ folder:

    python3 test_network.py
  

• Ensure that the folder contains valid gesture images.

🛠️ Troubleshooting

- Ensure correct paths for dataset, images, and model files.

- Use clear, well-lit hand gestures for better recognition.

For low accuracy:
        
        Increase the dataset size
        
        Improve image quality
        
        Tune network hyperparameters or retrain

📚 Citation If you use this project in your work, please cite:

  Malavika Suresh, Avigyan Sinha, Aneesh R P,
  “Real-Time Hand Gesture Recognition Using Deep Learning”,
  International Journal of Innovations and Implementations in Engineering (ISSN 2454-3489), 2019, Vol. 1.



  @article{malavika2019gesture,
    title={Real-Time Hand Gesture Recognition Using Deep Learning},
    author={Suresh, Malavika and Sinha, Avigyan and R P, Aneesh},
    journal={International Journal of Innovations and Implementations in Engineering},
    volume={1},
    year={2019},
    issn={2454-3489}
  }