Project for Signal, Image and Video at the University of Trento A.Y.2023/2024
Developed by:
De Martini Davide
Mascherin Matteo
For this project, our goal is to develop software capable of real-time traffic detection. Given a video input (e.g. a video of a security camera of a highway), the software will identify the street, crop it, and pinpoint vehicles by highlighting their bounding boxes along with their respective directions. To simulate a real time video flow, an HTTP server takes the input video, process it and streams the resulting frames to the connected clients. The video stream quality is selected by the client according to the network capabilities.
Desktop/traffic_detection
├── assets
│ ├── moving_cam.mp4
│ ├── video1.mp4
│ ├── video2.mp4
│ └── video3.mp4
├── config.py
├── counting_vehicles.py
├── http_streaming_server.py
├── README.md
├── requirements.txt
├── sockets
│ ├── receiver.py
│ └── sender.py
└── templates
├── index.html
└── video_viewer.html
http_streaming_server.py is the file for starting the webserver. This will expose two API for the client to connect to the server and receive the video stream.
counting_vehicles.py is the file to run directly a simulation of the traffic detection. It will read the video from the assets folder and process it. It also contains the code for the traffic detection used in the server.
In order to run the project you'll need to clone it and install the requirements. We suggest you to create a virtual environment
-
Clone it
git clone https://github.com/davidedema/traffic_detection
-
Create the virtual environment where you want, activate it and install the dependencies
cd path/of/the/project python -m venv /name/of/virtual/env source name/bin/activate pip install -r requirements.txt
The project could be runned in two different ways:
-
Through web server
python http_streaming_server.pyAfter running the server, the url is shown directly in the terminal. You can connect to the server by typing the url in the browser. If you want to watch the stream from a media player such as VLC, you can simply execute the following command:
vlc servel_url/stream_video
-
Running directly:
python traffic_detection.py