Antok ka na ba? Detecting drowsiness in video feeds
The paper aims to create a proof of concept for the application of deep learning in detecting drowsiness in video feeds. This is for the purpose of quantifying energy levels in videos of different individuals for different uses such as aiding in the maintenance of interest in online classes, ensurin...
Saved in:
Main Author: | |
---|---|
Format: | text |
Published: |
Animo Repository
2021
|
Subjects: | |
Online Access: | https://animorepository.dlsu.edu.ph/faculty_research/11126 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | De La Salle University |
id |
oai:animorepository.dlsu.edu.ph:faculty_research-11605 |
---|---|
record_format |
eprints |
spelling |
oai:animorepository.dlsu.edu.ph:faculty_research-116052023-10-25T23:52:12Z Antok ka na ba? Detecting drowsiness in video feeds Delgado, Kevynn P. The paper aims to create a proof of concept for the application of deep learning in detecting drowsiness in video feeds. This is for the purpose of quantifying energy levels in videos of different individuals for different uses such as aiding in the maintenance of interest in online classes, ensuring focus in attention critical jobs, as well as minimizing driver accidents caused by sleepiness. Synthetici video recordings collected were converted to frames at a rate of two frames per second using the OpenCV library in Python, treating this as a time series problem where each frame is a point in time. Haar Cascade and Local Binary Fitting were implemented on each frame, detecting the area of the face and recognition of the landmarks, respectively. With a deep learning architecture utilizing a Long Short-Term Memory (LSTM), an accuracy of 85% was achieved on the synthetic data. 2021-01-01T08:00:00Z text https://animorepository.dlsu.edu.ph/faculty_research/11126 Faculty Research Work Animo Repository Pattern recognition systems Deep learning (Machine learning) Neural networks (Computer science) Computer Sciences |
institution |
De La Salle University |
building |
De La Salle University Library |
continent |
Asia |
country |
Philippines Philippines |
content_provider |
De La Salle University Library |
collection |
DLSU Institutional Repository |
topic |
Pattern recognition systems Deep learning (Machine learning) Neural networks (Computer science) Computer Sciences |
spellingShingle |
Pattern recognition systems Deep learning (Machine learning) Neural networks (Computer science) Computer Sciences Delgado, Kevynn P. Antok ka na ba? Detecting drowsiness in video feeds |
description |
The paper aims to create a proof of concept for the application of deep learning in detecting drowsiness in video feeds. This is for the purpose of quantifying energy levels in videos of different individuals for different uses such as aiding in the maintenance of interest in online classes, ensuring focus in attention critical jobs, as well as minimizing driver accidents caused by sleepiness. Synthetici video recordings collected were converted to frames at a rate of two frames per second using the OpenCV library in Python, treating this as a time series problem where each frame is a point in time. Haar Cascade and Local Binary Fitting were implemented on each frame, detecting the area of the face and recognition of the landmarks, respectively. With a deep learning architecture utilizing a Long Short-Term Memory (LSTM), an accuracy of 85% was achieved on the synthetic data. |
format |
text |
author |
Delgado, Kevynn P. |
author_facet |
Delgado, Kevynn P. |
author_sort |
Delgado, Kevynn P. |
title |
Antok ka na ba? Detecting drowsiness in video feeds |
title_short |
Antok ka na ba? Detecting drowsiness in video feeds |
title_full |
Antok ka na ba? Detecting drowsiness in video feeds |
title_fullStr |
Antok ka na ba? Detecting drowsiness in video feeds |
title_full_unstemmed |
Antok ka na ba? Detecting drowsiness in video feeds |
title_sort |
antok ka na ba? detecting drowsiness in video feeds |
publisher |
Animo Repository |
publishDate |
2021 |
url |
https://animorepository.dlsu.edu.ph/faculty_research/11126 |
_version_ |
1781418224945987584 |