Understanding the Factors Underlying Variation in Detection Errors of Video- and Thermal-Imaging Cameras

Abstract

Video-imaging technology has long been used for vehicle detection at signalized intersections. This technology provides many advantages over conventional inductive loops. Video-imaging cameras are easier and safer to install and maintain; the cameras cause no damage to the pavement; and detection zones can be adjusted or redrawn when travel lanes are realigned or reassigned as a result of widening or remarking of the pavement. Despite the advantages, video-imaging cameras are generally susceptible to varying lighting conditions. Thermal-imaging cameras produce images on the basis of the difference in thermal energy emitted by objects and do not rely on visible light. Thus, the cameras are more robust in various lighting conditions. In this paper, both video-imaging and thermal-imaging technologies were evaluated for stop bar vehicle detection at signalized intersections. The detection errors were correlated with concurrent weather and environmental conditions. The performance of both types of cameras appears to be affected by wind and various weather events. In particular, uneven shade appears to be an issue for the video-imaging camera but not for the thermal-imaging camera. The video-imaging camera seems to be more susceptible to false calls at night, likely caused by vehicle headlights. A higher chance of missed calls and calls stuck on was experienced for the thermal-imaging camera at night.