Camera Based Traffic Light Control System

This book introduces a solution to one of the society problems (monitoring of roads area coverage by vehicles for avoidance of traffic congestion) by using a camera. The main idea is to simulate the traffic policeman’s actions in controlling the traffic light duration (timer) and the traffic pass permission (green light) through analyzing images sequences to determine the road occupation by vehicles (road density). The traffic duration is controlled according to the road density percentage; which is determined using an effective proposed metric for denasity calculation. The traffic pass permission is given to the road with the highest score first or the road that exceed waiting limits. The road score depends on the road priority, waiting time, and the calculated density. Empty road will not take the traffic pass permission as in the traditional system. The Main solved problems are: Camera vibrations, objects detection from its complex background, illumination changes (during day time, within the same frame, and from one fame to another), shadows elimination, background updating. The work also presents effective metrics to test the reference block uniqueness and high contrast.

Saja Sabeeh Azeezis an assistant lecturer in the University of Technology, Baghdad, Iraq. Her B.Sc. (2008) in computer science was obtained from the University of Technology, Baghdad, Iraq, and her M.Sc. (2013) in computer science was obtained from Baghdad University, Baghdad, Iraq. Research interests: image processing, security, and AI.