| Title |
Road Traffic Outlier Detection Technique Based On Linear Regression |
| ID_Doc |
46758 |
| Authors |
Mondal M.A.; Rehena Z. |
| Year |
2020 |
| Published |
Procedia Computer Science, 171 |
| DOI |
http://dx.doi.org/10.1016/j.procs.2020.04.276 |
| Abstract |
Road traffic anomaly detection is a very important aspect of intelligent traffic management system. Traffic anomaly may arise due to several reasons like unusual traffic incidents and malfunctioning of sensors deployed over the road network to capture traffic information. Unusual traffic incident includes road accident, road blockage due to construction, any major events and so on. For smooth mobility of the citizens, it is very important to identify these kinds of scenarios with minimum delay so that traffic management authority can take proper measures. This paper proposes a technique based on statistical model which identifies the temporal outliers in the road traffic. Z-score and linear regression model are two statistical models have been used in combination for detection of temporal outliers. The proposed technique can be used to detect unusual traffic incident or sensors failure. © 2020 The Authors. Published by Elsevier B.V. |
| Author Keywords |
automatic incident detection; linear regression model; smart city; traffic outlier; z-score |
