
Blinking Eyes Detection to Monitor Drowsy Drivers Due to Fatigue Using MATLAB Cascade Object Detector
Author(s) -
Zulfikri Paidi,
Nurul Awanis Noor Shaarin,
Nurzaid Muhd Zain,
Mahmod Othman
Publication year - 2021
Publication title -
journal of computing research and innovation
Language(s) - English
Resource type - Journals
ISSN - 2600-8793
DOI - 10.24191/jcrinn.v6i4.244
Subject(s) - haar like features , computer vision , matlab , artificial intelligence , computer science , detector , simulation , face (sociological concept) , pattern recognition (psychology) , face detection , facial recognition system , telecommunications , operating system , social science , sociology
Road accidents are incidents that should be avoided. One of the contributing factors of road accidents occurrence is drowsiness while driving due to fatigue. In this project, fatigue and drowsiness of a person can be detected by looking at the eye area. Drowsy situations are dangerous especially when driving a vehicle over long distances. When a person starts to feel drowsy, the eyes will start to blink more frequently. This characteristic can be used to monitor a driver’s fitness level. In this project the Viola-Jones algorithm using MATLAB cascade object detector was used to detect the presence of blinking eyes. The algorithm is consisted of three phases which are, selection of Haar-like characteristics, integrating a picture into the whole, and Classifiers in a cascading fashion. A number of 29 samples images consisting of the condition fatigue face and non-fatigue face was used. The obtained results obtained were calculated based on the accuracy value of the detected blinking eye. The result shows that both the total Area Under the Curve (AUC) values for faces with fatigue situations and non-fatigue situations are above acceptable values which is 0.5. This indicates both classifications are acceptable and can be used to detect the presence of blinking eyes which represent drowsiness.