In the initial phase of our project, we implemented the Haar Cascade algorithm for real-time facial detection, leveraging the capabilities of OpenCV for efficient feature tracking and dlib for precise facial landmark detection. Additionally, the iBUG 300-W datasheet, featuring 68 (x, y) coordinates for facial landmarks, acted as our guiding reference.

Moreover, we engineered an eye state detection algorithm that uses horizontal and vertical distances between eye coordinates, employing machine learning techniques to discover with remarkable accuracy whether the eyes were open or closed.

DrowseDetect Pro represented the final chapter of my university journey as it served as my senior project while pursuing my computer engineering degree. The project involved well-studied design, development, and extensive testing, gathering positive result feedback from various lighting scenarios, for different individuals, and for special cases like drivers wearing hats and glasses.

Attached, you will find the Software Architecture Flowchart Diagram, followed by a visual representation of the iBUG 300-W datasheet. Additionally, there is an early-stage photo capturing the project's inception.