Emotion Detection using CNN

This project focuses on facial emotion recognition using a Convolutional Neural Network (CNN) trained on grayscale images of size 48x48. It classifies faces into seven different emotions.

📁 Dataset Structure

Make sure your dataset is structured as follows:

dataset/
├── train/
│   ├── angry/
│   ├── disgust/
│   ├── fear/
│   ├── happy/
│   ├── neutral/
│   ├── sad/
│   └── surprise/
└── test/
    ├── angry/
    ├── disgust/
    ├── fear/
    ├── happy/
    ├── neutral/
    ├── sad/
    └── surprise/

Images must be 48x48 pixels in grayscale format.

🧠 Model Architecture

• Convolutional layers with increasing filters (32 → 256)
• Batch Normalization after each convolution
• MaxPooling layers after each block
• Dropout layers to prevent overfitting
• Flatten followed by two Dense layers
• Output layer with Softmax for 7 classes

🛠️ Techniques Used

• Data Augmentation: Rotation, zoom, shift, flip for better generalization
• Class Weights: Applied to balance underrepresented classes
• Callbacks:
  • EarlyStopping
  • ReduceLROnPlateau
  • ModelCheckpoint

🔧 Requirements

Install dependencies using:

pip install tensorflow matplotlib seaborn scikit-learn

🚀 Running the Model

Training

Ensure you're in the correct working directory and run:

python train_model.py

Output Files

• emotion_model.h5: Final saved model
• best_emotion_model.h5: Best performing model based on validation loss

📊 Evaluation

The script generates:

• Confusion Matrix
• Classification Report (Precision, Recall, F1-score)
• Final test accuracy, precision, and recall

📷 Results Visualization

Confusion matrix heatmap(emotion_model.h5):

Classification Report(emotion_model.h5):

---

📌 Notes

• Make sure image size and channels match the model input (48, 48, 1).
• Adjust class weights depending on your dataset’s class imbalance.
• You can visualize training curves by extending the history object.

---

📁 File Structure

emotion_project/
├── images/
│   ├── EmoModel_ConfusionMatrix.jpg
│   └── EmoModel_ClassificationReport.jpg
├── models/
│   ├── emotion_model.h5
│   └── best_emotion_model.h5
├── augment_disgust.py
├── evaluate_model.py
├── predict_emotion.py
├── README.md
└── train_model.py