Bus lanes are supposed to be utilised solely by buses. But often, we see unauthorised vehicles utilizing these lanes, as a way to bypass traffic. This project demonstrates that we can use Deepstream and Computer Vision to detect such vehicles in a Video analytics pipeline.
-
Installing System Dependencies for DeepStream
sudo apt install \ libssl1.0.0 \ libgstreamer1.0-0 \ gstreamer1.0-tools \ gstreamer1.0-plugins-good \ gstreamer1.0-plugins-bad \ gstreamer1.0-plugins-ugly \ gstreamer1.0-libav \ libgstrtspserver-1.0-0 \ libjansson4=2.11-1 uuid \ uuid-dev
-
DeepStream installation on Jetson Device Download the deb file under - DeepStream 5.1 for Jetson from here
sudo apt-get install ./deepstream-5.1_5.1.0-1_amd64.deb
-
Clone the repo
git clone https://github.com/Shalom-P/wrong_lane.git cd wrong_lane
-
Running the application
- Firstly we need to build the application
make
- Put the absolute file path of the video source as an argument to the binary (The demonstration is using the sample.mp4 video as the ROI is defined for that particular video)
./bus_lane file:///your_working_directory/wrong_lane/vids/sample.mp4
- Firstly we need to build the application