An Architecture of Video Management System for Intelligent Transportation and Traffic Control Management

Transportation and traffic control managers have to deal with thousands of vehicles running through streets and causing traffic congestion daily. In addition, other issues such as increased travel time, incidents, anomalies, environmental pollution, etc. are becoming a matter of concern, especially in metropolises. In the last few decades, the number of closedcircuit TVs (CCTV) has been increasing, in order to provide more efficient security and safety in cities. As these systems become larger, effectively observing all cameras, detecting target events, and extracting useful information in a timely manner become a challenge especially in public and crowded places. Whilst videos are the rich sources of traffic analytics parameters and must not be missed. In this paper, we discuss the applications of CCTV video processing in transportation and traffic control management by using deep learning algorithms as the best solution to process such huge and unstructured data. Then, as the first step toward using data in decision-making is providing an integrated system to receive and manage data properly, an integrated architecture of video management system (VMS) is introduced in order to facilitate video processing to boost the efficiency of decision-making. Finally, we compare our VMS with well-known and existing VMSs.

Keywords:

video processing , video management systems , intelligent transportation and traffic management , CCTV