Helmet Detection Based on Cascade Classifier and Adaptive Boosting

Abstract

The increasing number of traffic accidents caused by motorcyclists not wearing helmets has led to an increase in the number of studies related to road safety surveillance. The research system used is an automatic system to detect whether the motorcyclist is wearing a helmet or not. Many studies use image processing systems, deep learning and computer vision. In this research, Cascade Classifier and Adaptive Boosting have been implemented for the process of identifying motorcycle riders with helmets and without helmets. The number of datasets used is 500 datasets with labels on the image of the driver with a helmet and the image of the driver without a helmet. Based on the test results, an accuracy of 90% has been obtained