Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Rahma Tiara Puteri
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Deteksi Kantuk Menggunakan Kombinasi Haar Cascade dan Convolutional Neural Network Rahma Tiara Puteri; Fitri Utaminingrum
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 3 (2020): Maret 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (283.227 KB)

Abstract

The number of traffic accidents is getting more numerous, especially in the time approaching the Lebaran season where people will go home to their villages. Based on data in 2017, there were 73 accidents in the six days before Eid, and when compared to 2016 there were 63 accidents, the number increased quite a lot by 16%. The main factor causing the accident is due to fatigue or drowsiness, because most events are at 21.00-24.00 then followed at 03.00-06.00.. Therefore, we need a system that can detect the state of the driver when he is tired or sleepy. This research developed a drowsiness detection system using Haar Cascade and Convolutional Neural Network. Input to the system is obtained from the Logitech C310 Webcam which will capture an image of a face. The main processing system uses Intel NUC5i7RYH which is used for image processing. The output of the system is the sleepy warning on the monitor when the driver is sleepy and there is an alarm sound for warning. The average accuracy of the system for face detection using Haar Cascade is 100%, the average accuracy for detecting open and closed eyes at a distance of 30-50 cm is 97.23% and the average accuracy for the detection of drowsiness is 97.23%. This system has an average computing time of 0.2075 s which will make it easy to detect drowsiness quickly.
Co-Authors Utaminingrum, Fitri