Emotion-Detection-System

This repository contains the implementation of an emotion recognition model using the VGG16 architecture.The project involves fine-tuning a pre-trained VGG16 model for emotion classification using the FER2013 dataset combined with additional custom facial emotion datasets, and the necessary code is provided.

Contents of the Repository

1. Vggmain.ipynb: The main notebook for setting up the VGG16 model, training, and evaluating it on the dataset.
2. try 2 vgg16 fine tune.ipynb: A notebook showcasing fine-tuning techniques applied to the VGG16 model to improve performance.

Key Features

• Fine-tuning of the VGG16 model using the FER2013 dataset along with custom emotion image datasets.
• Includes trained model weights to save training time.
• Reproducible code for training and evaluation.

Requirements

To run the notebooks and use the model, you need the following:

• Python 3.7 or later
• TensorFlow (>=2.0)
• Keras
• Jupyter Notebook
• NumPy
• Pandas
• Matplotlib

Install the dependencies using:

pip install tensorflow keras numpy pandas matplotlib

Usage

1. Clone the repository:

   git clone https://github.com/your-username/emotion-recognition-vgg16.git
   cd emotion-recognition-vgg16

2. Open the notebooks in Jupyter:

   jupyter notebook Vggmain.ipynb

3. To fine-tune the model, open and run the try 2 vgg16 fine tune.ipynb notebook.

Model Details

• Architecture: VGG16 (Pre-trained on ImageNet)
• Dataset: FER2013 dataset combined with additional custom facial emotion dataset (not included; you can replace it with your own dataset).
• Output: Emotion classes such as Happy, Sad, Angry, Neutral, etc.

Results

The fine-tuned model achieved the following accuracy on the test dataset:

• Accuracy: 67%
• Loss: 0.87

How to Use the Pre-Trained Model

You can load the model weights as follows:

from tensorflow.keras.models import load_model


## Contribution

Feel free to contribute to this repository by creating issues or submitting pull requests. Contributions can include:
- Improving the documentation
- Adding more datasets
- Enhancing the model architecture
- Fine-tuning techniques

## License

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

## Author

[Sriyanka Baral]  
LinkedIn: https://www.linkedin.com/in/sriyanka-baral-382795282/