Video analytics for multi-camera traffic surveillance
Dongyu Ang, Yao Shen, Prakash Duraisamy · 2009
A low-cost Video Image Detection Systems (VIDS) is introduced. Using video analytics, the system can count the number of vehicles making a left (or right) turn at an unseen intersection plus collect statistics on other traffic conditions. Other functionality can be added. Two cameras with non-overlapping views were used as the information source. Between them was situated a "T" intersection. Comparing the automated data collected from the two cameras' video with manually generated truth data, no errors were found over five minutes of video.