Improved traffic light based on image processing

In modern era, roadways are the most important asset for transportation in the world. Vehicles usage continues increase from time to time. Nowadays, traffic jam is always happen by long red light time delays to switch and also the wasting time of green light at empty road. So, traffic control system must need for an improvement and optimized. Microcontrollers and microprocessors using hardware that has been prescribed, that work from the program that does not efficiency for real-time modification of the present traffic light controller. An improved traffic light system detect and count the existence or absence of the vehicles and reacts with that by using an image processing techniques of image detection to control traffic in the country at a crossroad. When cars has count at a fixed number, the green light turn on with the appropriate time that has been set according to the number of car detected. In terms of efficiency and convenient, it is an improvement for the existing traffic light control system. A tested with a real-time computer simulation, car detection would be done by using method Haar Cascade to count number of cars and identify existing of cars. Python is use to program and can be implement by using Open Computational Vision (OpenCV) as a library image to simulate this project. With this project, it can get rid of time delay by reduce automatically traffic and controlling signal.