A YOLOv8-based object detection system that supports inference on images, videos, and live camera feeds with optional server integration for real-time alerts.

• 🎯 Object Detection - Detect objects using custom-trained YOLO models
• 📊 Congestion Analysis - Calculate real-time congestion levels
• 🎥 Multi-source Input - Process images, videos, folders, or live camera feeds
• 🖥️ Server Integration - Send detection alerts to a backend server

seds/
├── inference.py        # Main CLI script for batch inference
├── client.py           # Real-time detection client with server integration
├── congestion.py       # Congestion analysis module
├── yolo.py             # Core YOLO utility functions
├── config.py           # Model configuration settings
├── model_info.py       # Utility to inspect model details
├── requirements.txt    # Python dependencies
├── your_model.pt       # Your trained YOLO model
├── test_images/        # Sample test images
└── output/             # Inference results output directory

• Python 3.8+
• CUDA-compatible GPU (optional, for faster inference)

# Clone the repository
git clone https://github.com/TanmayBansode/seds.git
cd seds

# Create virtual environment (recommended)
python -m venv venv
venv\Scripts\activate  # Windows
# source venv/bin/activate  # Linux/macOS

# Install dependencies
pip install -r requirements.txt

Edit config.py to specify your trained model:

model_file = "your_model.pt"

# Single image
python inference.py -i path/to/image.jpg

# Video
python inference.py -v path/to/video.mp4

# Folder of images/videos
python inference.py -f path/to/folder/

Results are saved to output/ with labeled media and detection details in .txt files.

python client.py

Processes live camera feed with real-time detection and sends alerts to a configured server.

python congestion.py --source path/to/video.mp4

Calculates congestion levels based on detected object coverage.

python model_info.py

Displays model architecture, class names, and hyperparameters.

Images: jpg, jpeg, png, bmp, webp, tif, tiff

Videos: mp4, avi, mov, mkv, webm