Comprehensive Classification System for Online Learning

This project provides a comprehensive classification system designed for online learning environments. The system uses advanced deep learning models for analyzing images and videos, extracting frames, and processing video data in real-time. It can detect and classify various behaviors and states like engagement, distraction, and fatigue, providing insights into student behavior.

Features

• Normal Image Processing: Upload normal images to detect and classify behaviors.
• Frame Extraction from Videos: Extract and analyze frames from uploaded videos.
• Real-Time Video Processing: Process and classify uploaded videos in real-time.

Technologies Used

• Python: Programming language used for implementing the system.
• Streamlit: Framework used for building the web application.
• YOLO (You Only Look Once): Object detection model for detecting objects in images and videos.
• TensorFlow: Framework used for loading and running the classification model.
• OpenCV: Library used for image and video processing.

Installation

1. Clone the repository:

   git clone https://github.com/abdellatif-laghjaj/classification-system-for-online-learning.git

2. Install the required packages:

    pip install -r requirements.txt

3. Run the Streamlit application:

   streamlit run app_classification.py --server.runOnSave true

4. Open the Streamlit app in your browser:

   http://localhost:8501
   

Aknowledgements

• Streamlit: https://streamlit.io/
• YOLO (You Only Look Once): https://pjreddie.com/darknet/yolo/
• TensorFlow: https://www.tensorflow.org/
• OpenCV: https://opencv.org/
• Python: https://www.python.org/

Author

• Abdellatif Laghjaj - abdellatif-laghjaj