Yawning, a physiological phenomenon is a potential indicator of alertness and drowsiness. Recognizing the impracticality of manual monitoring, the study proposes an innovative vision-based machine-learning algorithm for the detection of yawning. Grounded in OpenCV, Dlib, and Python, this automated system analyzes facial features in real time to detect yawning patterns, offering practical applications in contexts such as driving safety and classroom environments. Surveys highlight correlations between excessive yawning and sleep-related issues, motivating the need for precise detection methods. This paper explores the innovatory integration of Convolutional Neural Networks (CNNs) for automated yawning detection, revolutionizing the understanding of this complex physiological phenomenon. Departing from traditional manual monitoring, our proposed vision-based machine learning algorithm leverages the power of CNNs, OpenCV, and Dlib to enhance the yawning detecting system's precision and effectiveness in real time. The application extends to diverse scenarios, including driving safety and classroom fatigue assessment. The system outputs binary classifications, signaling instances of yawning or non-yawning.