Detection of Helmet Use in Motorcycle Drivers Using Convolutional Neural Network

The lack of helmet use in motorcyclists is one the main risk factors with severe consequences traffic accidents. Wearing a certified motorcycle can reduce head injuries by 69% and fatalities 42%. At present there are systems that detect very precise way, however they not robust enough to guarantee safe journey, why proposed an intelligent model for detecting real time using training images camera mounted on motorcycle, convolutional neural networks allow constant monitoring region interest identify helmet. As result, was obtained capable identifying when used or objective manner while user making performance 97.24%. Thus, it possible conclude this new safety perspective provides first approach generation preventive help accident rates these means transport. future work, improve different may violate detection.