Visual Analytics of Traffic Congestion Propagation Path with Large Scale Camera Data

Abstract

Congestion analysis is essential to traffic control, especially in crowded urban road network. The recent traffic forecasting methods can provide travelers and traffic managers with early congestion warning, yet unable to reveal the relationship of congestion roads. This paper presented a congestion propagation path estimation method based on greedy algorithm to quickly extract these congestion relationships for visual analytics. The data from traffic cameras are applied to build the propagation network based on a directed weighted graph. It describes the process of congestion spreading among different segments. According to this network, congestion propagation path predicts the process of congestion spreading between different segments. In our visual design, it is applied to demonstrate the segments that will be influenced by the congested road. This is helpful for traffic managers to make effective and efficient decisions. The experimental result shows that our method achieves high accuracy thus prove the effective for the congestion propagation method.