😊 Facial Expression Recognition with GUI

A Python project to detect facial emotions from images using a simple machine learning model (Decision Tree Classifier) and display results in a user-friendly Tkinter GUI.

---

📂 Project Overview

This project uses grayscale facial images to train a model that classifies them into one of the 7 basic human emotions:

• 😠 Anger
• 😒 Contempt
• 🤢 Disgust
• 😨 Fear
• 😀 Happiness
• 😢 Sadness
• 😲 Surprise

It provides a GUI for uploading a facial image and viewing the predicted emotion.

---

📁 Dataset Structure

Place your emotion image folders in the ml project directory as follows:

ml project/
├── anger/
├── contempt/
├── disgust/
├── fear/
├── happiness/
├── sadness/
└── surprise/

Each folder should contain .jpg/.png/etc. images corresponding to that emotion.

---

🔧 Requirements

Install the following Python libraries:

pip install opencv-python scikit-learn numpy joblib Pillow

---

🚀 How to Run

1. Ensure your dataset is placed in the proper folder structure.
2. Run the Python script:

python main.py

3. A GUI window will open. Click "Upload Image" to test an image from your system.

---

📌 Features

• Decision Tree Classifier trained on 48x48 grayscale facial images

• Preprocessing includes:

  • Grayscale conversion
  • Resizing
  • Flattening images for training

• Uses Tkinter GUI for easy image upload and prediction

• Automatically saves the trained model as emotion_model.pkl

• Handles non-image and unreadable files gracefully

---

💻 GUI Preview

• Upload any image containing a face
• The GUI displays the image and the predicted emotion

---

📊 Model Evaluation

At the end of training, the console will display:

• ✅ Accuracy Score
• 📋 Classification Report

Example:

Accuracy: 0.81
Classification Report:
              precision    recall  f1-score   support
   happiness       0.89      0.85      0.87        52
       anger       0.78      0.82      0.80        50
      ...

---

🧠 Emotions Dictionary

emotions = {
    0: 'anger',
    1: 'contempt',
    2: 'disgust',
    3: 'fear',
    4: 'happiness',
    5: 'sadness',
    6: 'surprise'
}

---

🛠 Future Improvements

• Upgrade model to Convolutional Neural Network (CNN) for better accuracy
• Add webcam live detection
• Include data augmentation
• Save predictions or logs to a file

---

👤 Author

Aditya10045 Made with ❤️ and Python

---

📜 License

This project is open-source and free to use for educational or personal purposes.