<![CDATA[Traffic accident factors are caused by 3 factors such as human negligence, the state of the vehicle and also environmental factors. According to WHO (World Health Organization) the use of helmets for motorcycle riders can reduce the risk of death by up to 40%. The use of helmets as safety for motorcycle riders is still considered not so important that it is ignored, therefore motorcycle accidents, especially motorcyclists who do not use helmets and the level of traffic violations are high. The fatigue factor for the police in monitoring traffic causes ignoring motorists who violate by not wearing a helmet while driving. Therefore helmet detection for motorcycle riders is very important, namely using technology to obtain information on motorcycle riders who violate the rules. Based on these factors, after knowing the causes of motorcycle accidents in traffic, one solution can be reduced, namely using computer vision for helmet detection as motorcycle riders' safety in driving and facilitating the work of police officers in guarding traffic by way of notification. notification of traffic violations of motorists who are not wearing helmets with buzzer alerts. The author proposes the CNN (Convolutional Neural Network) method as a detection of motorcyclists who violate traffic such as not wearing a helmet, thereby reducing traffic accidents. The results of the tests that have been carried out by the system can detect objects of people not wearing helmets with an accuracy of 90% using the confusion matrix on the test results.]]>
Copyrights © 2021
