According to the road traffic injuries fact sheet from the World Health Organization, approximately 1.3 million people die each year because of road traffic crashes. There are many risk factors and one of the most important is distracted driving, especially caused by using a mobile phone. The fact sheet shows that drivers using mobile phones are approximately 4 times more likely to be involved in a crash than drivers not using a mobile phone. Apart from this, dangerous driving behaviors and driver emotions are growing concerns for road safety. This project aims to develop a model for detecting dangerous driving behaviors and analyzing drivers’ emotions in order to avoid and minimize traffic accidents. We focus on detecting several distractive behaviors, bad driving practices, and abnormal emotions. We carry out experiments to evaluate the system performance through the integration of our system with existing infrastructure. To the best of our knowledge, we design and develop the first small and portable device to collect data from drivers, then the collected data is transferred to the cloud for further analysis. The trial service is already available for drivers for a few months in Hong Kong.