Author: Jan Haslik (@janhaslik)
TrafficWatch is a real-time traffic monitoring application that utilizes object detection to count vehicles, such as cars and trucks, across multiple cameras to deliver real-time traffic analytics. This system integrates Kafka for streaming data, Java Spring for backend services, React with WebSocket for real-time display, MongoDB for data storage, and OpenCV, Keras, Tensorflow for computer vision tasks.
Kaggle Dataset for the Object Detection Model: Click Link
🚀 Features
- Object Detection (Coming Soon): Utilizes OpenCV, Keras, and Tensorflow to detect and count vehicles in camera feeds.
- Kafka Streaming: Streams real-time live camera feed & traffic data for efficient and scalable data processing.
- Dynamic Web Interface: Built with React and WebSocket to display real-time traffic analytics and camera feeds.
- Data Storage: Stores traffic data in MongoDB for historical analysis and reporting.
Here are some screenshots of the TrafficWatch user interface:
