Image Processing Application to Detect Driver Drowsiness using Raspberry PI kit

Abstract

In this project, we aim to find real-time responsive algorithms, optimal and simple solutions that provide high accuracy and meet practical needs. We simulate simple image processing problems, static image recognition, and direct recognition through the camera using Python programming. We also develop accurate eye state tracking in realtime on the Raspberry Pi 4 Kit. Face detection is performed using the AdaBoost classifier method based on Haar-like features. Eye recognition is achieved by marking the facial features using the Facial Landmarks algorithm, and then calculating the Euclidean distance between the two eye folds to detect the eye state and identify drowsiness.