NORMA eResearch @NCI Library

Multi-label Image Classification to Detect Air Traffic Controllers’ Drowsiness Using Facial Features

Saini, Nevin (2019) Multi-label Image Classification to Detect Air Traffic Controllers’ Drowsiness Using Facial Features. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (674kB) | Preview
[thumbnail of Configuration manual]
Preview
PDF (Configuration manual)
Download (1MB) | Preview

Abstract

With the increase in globalization, there is a rise in the frequency of airplanes activities and air traffic request causing immense financial profits to the air carrier. To maintain the same progression, it is necessary to consider the safety of passengers that require highly skilled Air Traffic Controller who can clearly communicate with the pilots and sustain the safety of crew and passengers. Also, working as an ATC (Air Traffic Controller) involves extreme pressure due to the safety of passengers and over load of work which leads to drowsiness and fatigue among the staff affecting the quality of work and endangering the life of travellers. This study intends to design a drowsiness detection model to overcome the above issue by using numerous modeling techniques like Convolutional Neural Network, Support Vector Machine, K-Nearest Neighbor and some ensemble techniques like XGBoost and Random Forest which can detect and classify facial features like Closed eyes, Yawning and Open eyes of an ATC. Two different models: Baseline and Tuned/Improved model after hyperparameter tuning are designed to improve the performance of an baseline algorithm and finally, these techniques are assessed on the basis of performance measures like Accuracy, Precision, Recall and F1 Score on the dataset used in this study and, it is observed that Convolutional Neural Network has outperformed other machine learning models and obtained an accuracy of 98.6% without overfitting.
Keywords: Drowsiness Detection System, Aviation, Convolutional Neural Network, Hyperparameter Tuning

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
T Technology > TL Motor vehicles. Aeronautics. Astronautics
H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Aviation Industry
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 15 Jun 2020 10:54
Last Modified: 15 Jun 2020 10:54
URI: https://norma.ncirl.ie/id/eprint/4280

Actions (login required)

View Item View Item