Research into methods for determining driver drowsiness
Abstract
This paper discusses the expansion of the capabilities of vehicle driver drowsiness detection based on surveillance camera signals. To this end, features associated with driver blinking and yawning are extracted in experiments. Based on this dataset, a feature selection method is developed to detect driver state. The results obtained will help in the future to develop reliable driver drowsiness monitoring systems to prevent fatigue-related accidents. Modern driver fatigue monitoring systems that analyze driving performance to collect driver state information are also discussed. Such systems are capable of detecting drowsiness signs by evaluating steering behavior and warning the driver when drowsiness reaches a critical level. However, such systems do not have access to direct driver state data.