According to the most recent studies published by the World Health Organization (WHO) in 2013, it is estimated that 1.25 million people die as a result of traffic crashes. Many of them are caused by what it is known as inattention, whose main contributing factors are both distraction and drowsiness. Overall, it is estimated that inattention causes between 25% and 75% of the crashes and near-crashes. That is why this is a thoroughly studied field by the research community. The solutions to combat distraction and drowsiness, in particular, and inattention, in general, can be classified into three main categories, where computer vision has clearly become a non-obtrusive effective tool for the detection of both distraction and drowsiness. The reason for this Ph.D. thesis is the direct consequences of the abovementioned figures and its purpose is to provide mechanisms to help reduce driver inattention effects using computer vision techniques. More specifically, the aim of this thesis is to propose, build and validate an architecture based on the analysis of visual characteristics by using computer vision techniques and machine learning to detect both distraction and drowsiness in drivers. This architecture is designed to operate in vehicular environments, with a very low computational load and easily embeddable into devices with reduced computational capacities in order to deal with images in the different conditions prevailing in this type of environments. The proposed control system integrates several innovative elements in order to operate in a completely autonomous way for the robust detection of the main visual indicators characterizing the driver's both distraction and drowsiness. The architecture has been validated, firstly, with reference databases testing the different modules that compose it, and, secondly, with users in real environments, obtaining in both cases, promising results with a suitable computational load for the embedded devices in vehicle environments.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.