Implementation of Template Matching on Detection of Stop Line Violations

Abstract

Road marking is a sign located above the road surface that combines either a line or a symbol and provides information as visual guidance to road users. One example of road marking is the stop line behind the zebra crossing. Many violations of pedestrian rights are committed by motorized vehicles, especially because vehicles do not stop behind the stop line. This paper proposes a system to detect this type of traffic violation using the template matching method. This method is a technique in digital image processing to find small parts of the image that match the image template, so it can detect whether there is a traffic violation or not. This system uses the appropriate template images for each dataset (morning, afternoon, and evening). The system produces accuracy performances for the morning and afternoon dataset of about 100% and 78% respectively, and for the evening dataset is 70%. So, the overall average accuracy rate is 83%. The main factor affecting the overall accuracy performance is the inability to automatically extract the best template for the dataset from the morning and evening datasets.