Solitary of the main reason of traffic accident is drowsiness among drivers. All over the world, the number of deadly injuries and fatal deaths rises annually. Road accidents might be decreased by identifying the driver's sleepiness. A machine learning technique for detecting tiredness is described in this research. The areas of the driver's eyes that serve up as template for observing them in later frames are found using face detection. Lastly, tiredness is detected using the tracked eye's visuals to provide warning alarms. In order to prevent accidents caused by fatigued drivers, this paper offers a thorough assessment of current approaches and suggests a novel method for automatic sleepiness detection. Extensive techniques for facial recognition are required to accurately identify tiredness from facial expressions. Additionally, this study looks at how well a variety of algorithms like Support Vector Machine, Random Forest, Convolution Neural Network, and XG boost, might improve drowsiness detection systems. The possible ramifications of this innovative strategy for enhancing visibility for drivers and lowering fatigue-related traffic accidents are also covered.