Fatigue that causes signs of drowsiness in drivers is one of the main causes of traffic accidents. These signs manifest themselves in different levels of severity, and the higher the level, the greater the risk of accidents. This work presents a method based on artificial intelligence, machine learning, deep learning and computer vision to develop a system capable of identifying signs of drowsiness and issuing alerts proportional to the level of fatigue detected. Using the YOLO (You Only Look Once) algorithm, widely recognized for its effectiveness in real-time object detection, a model was developed to recognize signs of drowsiness in drivers. The model construction process included essential steps, such as image
collection and model training. After training, the model was subjected to tests, which showed its efficiency in detecting signs of fatigue and its results were evaluated through statistical metrics, verifying its accuracy in identifying the different levels of fatigue. Based on these signals, the system can alert the driver in cases of severe fatigue, acting as a preventive tool to increase traffic safety. Thus, the system contributes to the reduction of accidents related to drowsiness.
|
