Desarrollo de una aplicación móvil para la detección de la somnolencia de conductores de vehículos aplicando inteligencia artificial.

Abstract

Today, complex data processing and decision making are limited by the large amount of data generated by humans and machines. Machine learning, part of artificial intelligence, is essential for making complex decisions. Facial recognition has various practical applications, such as security, authentication, commerce, health, entertainment and marketing, including identifying unknown people and personalizing offers. In road safety, it could be used in vehicles to detect driver distractions, fatigue or drowsiness, often caused by long periods of driving, lack of rest, sleep deprivation, irregular schedules and sleep disorders, with potentially life-threatening consequences. For this reason, it is proposed to use artificial intelligence to detect signs of drowsiness in drivers and activate a preventive alarm. This improves road safety and reduces costs, benefiting all drivers, even in low-end vehicles. For the development of the mobile application, the Mobil-D methodology was used, where, in chapter II, the procedure carried out according to the phases that this methodology has is shown. The survey was used as the main technique and a Microsoft 365 form was used as an instrument, in which the level of satisfaction of the respondents with respect to the use of the mobile application was analyzed and appreciated. To measure the degree of satisfaction, the Likert scale immersed in the questions posed was used. Finally, after the completion of the prototype evaluation, chapter III shows the conclusions, recommendations and future work. Among the conclusions, the completion of the application stands out, which demonstrates the viability and effectiveness of using advanced tools such as artificial intelligence to address real challenges in society. Among the recommendations, we talk about detecting drowsiness in drivers using artificial intelligence and facial recognition. It is recommended to collect various data, use convolutional neural networks and train to detect subtle signs of drowsiness, such as eye closure or changes in blinking.