Kontrol Traffic Light Otomatis Berbasis Image Processing

Abstract

The traffic light is a very important part for the means of road transport, especially in the big cities. Increasing the number of vehicles currently causes increased density on the roads that can cause congestion. Therefore, a proper traffic light control system is required to reduce road congestion. The camera can be one of the alternatives to monitor the condition of road vehicle so that traffic light control can be more efficient. The image data has been obtained from the camera and then processed using a combination of image processing methods with background subtraction techniques. This technique serves as a comparison object image with a reference image so it can be known the number of white pixels on the image subtraction results. Based on the number of white pixels that have been obtained, it can be seen the percentage of vehicle queue length and vehicle density. The percentage of data that has been obtained, then sent to the microcontroller in order to control the duration of the flame of the green light at the traffic light in accordance with the queue length and vehicle density. Based on the research that has been done, the control of traffic light using image processing method combination of background subtraction technique so that it can be known the density and the length of the vehicle queue based on the image of the image processing until the accuracy of each 77.03% and 91.18%.