Your browser is unsupported

We recommend using the latest version of IE11, Edge, Chrome, Firefox or Safari.

ECE.15 – Computer Vision Based Intelligent Transportation System

Team Members Heading link

  • Ziliang Hong
  • Xiao Yang

Project Description Heading link

The rapid increase in the number of vehicles on our roads has put a significant strain on our transportation systems, leading to issues such as traffic congestion, air pollution, and safety hazards. To mitigate these problems, it’s essential to implement innovative solutions that can improve traffic flow and encourage the use of alternative and more efficient forms of transport that are better for the environment, city design, health, safety, and cost.

Our project aims to develop a traffic system that addresses these concerns by leveraging advanced technologies and cutting-edge techniques. To achieve this, we have implemented a deep learning-based approach for the detection of moving vehicles in real-time. We are utilizing the Yolo V5 S neural network, trained with Pytorch, which can quickly and accurately detect and track objects in video streams. We are also employing computer vision techniques to analyze the motion of detected objects like vehicles and pedestrians and also determine their speed and direction. By comparing the vector between the position of the same object in different time stamps, We figured out what direction the pedestrian, the vehicle might go.We can then use this information to optimize traffic light patterns and make traffic flow more efficiently.
To process the video stream and perform the necessary calculations, we are utilizing a Jetson Nano, a powerful single-board computer that is specifically designed for high-performance edge computing.
Once the video has been processed, the Jetson Nano will send the processed video and other control signals back to the server. Our server is built on Java Spring, a popular framework for building enterprise-grade web applications. The server will then host a web page that displays real-time information on traffic flow and other relevant data. Users can access this information on their devices, including desktop computers, laptops, tablets, and smartphones.