Traffic accidents claim thousands of lives every year, representing one of the most formidable challenges to public health and road safety. Data from Brazil’s Federal Highway Police (PRF) reveals that more than 25% of fatalities on federal highways are directly associated with critical behaviors—such as drivers falling asleep at the wheel, delayed reactions, or even a complete lack of response. This alarming reality reinforces the urgent need for innovative solutions to mitigate these risks. Among the emerging strategies, the application of computer vision for the early detection of drowsiness in drivers stands out as a promising approach. This technique enables the training of advanced models capable of extracting and analyzing essential features from images, such as facial recognition and the identification of regions of interest (for example, eyes and mouth), as well as distinguishing between specific alert states (such as eyes open versus closed, mouth open versus closed, and subtle head movements). However, the effectiveness of these models crucially depends on the quality and diversity of the data used. Obtaining images for the training and validation of these models proves to be a challenge, as many articles do not make available the image databases used in their studies. In this context, the main contribution of this article is a comprehensive catalog of image databases that consolidates and organizes essential resources for the development and validation of drowsiness monitoring systems.