Chapter

Databases in Networked Information Systems

Volume 7813 of the series Lecture Notes in Computer Science pp 150-160

Real-Time Traffic Video Analysis Using Intel Viewmont Coprocessor

  • Seon Ho KimAffiliated withIntegrated Media Systems Center, University of Southern California
  • , Junyuan ShiAffiliated withDepartment of Electrical Engineering, University of Southern California
  • , Abdullah AlfarrarjehAffiliated withDepartment of Computer Science, University of Southern California
  • , Daru XuAffiliated withDepartment of Electrical Engineering, University of Southern California
  • , Yuwei TanAffiliated withDepartment of Computer Science, University of Southern California
  • , Cyrus ShahabiAffiliated withIntegrated Media Systems Center, University of Southern CaliforniaDepartment of Electrical Engineering, University of Southern CaliforniaDepartment of Computer Science, University of Southern California

Abstract

Vision-based traffic flow analysis is getting more attention due to its non-intrusive nature. However, real-time video processing techniques are CPU-intensive so accuracy of extracted traffic flow data from such techniques may be sacrificed in practice. Moreover, the traffic measurements extracted from cameras have hardly been validated with real dataset due to the limited availability of real world traffic data. This study provides a case study to demonstrate the performance enhancement of vision-based traffic flow data extraction algorithm using a hardware device, Intel Viewmont video analytics coprocessor, and also to evaluate the accuracy of the extracted data by comparing them to real data from traffic loop detector sensors in Los Angeles County. Our experimental results show that comparable traffic flow data to existing sensor data can be obtained in a cost effective way with Viewmont hardware.

Keywords

Video Analysis Intel Viewmont Traffic Flow Data Inference