Title: A web-based visual analytics system for traffic intersection datasets

Authors: Ke Chen; Tania Banerjee; Xiaohui Huang; Zhaowen Ding; Venkata Sai Varanasi; Anand Rangarajan; Sanjay Ranka

Addresses: Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA ' Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA ' Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA ' Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA ' Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA ' Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA ' Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA

Abstract: The efficiency of a road network may be improved by making the traffic intersections more efficient. A smart traffic intersection is equipped with different sensors from which it is possible today to collect streaming data feed and run data analysis algorithms to identify potential inefficiencies, near-miss incidents, and anomalous traffic behaviour. In this paper, we present a visual analytics framework which may be used by traffic engineers to analyse the events and performance at an intersection. The tool ingests streaming videos collected from a fisheye camera, cleans the data, and runs analytics on it. The tool presented here has two modes: a streaming mode and a historical mode. The streaming mode may be used to analyse data close to real-time with a latency set by the user. In the historical mode, the user can run a variety of trend analysis on historical data.

Keywords: visual analytics; intersection traffic analysis; trajectory analysis; anomaly detection.

DOI: 10.1504/IJBDI.2021.118759

International Journal of Big Data Intelligence, 2021 Vol.8 No.1, pp.76 - 88

Received: 15 Sep 2020
Accepted: 01 Oct 2020

Published online: 25 Oct 2021 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article