Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Teknik Informatika (JUTIF)
Abdillah, Annas
Unknown Affiliation
Computer Science & IT

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

Found 1 Documents
Search

 1 
DESMOCAM (DETECTION SMOKING CAMERA): INTEGRATION OF IOT AND MACHINE LEARNING FOR ACTIVE SMOKER DETECTION TO SUPPORT SMART CITIES IN INDONESIA Abdillah, Annas; Nayu, Balqist Kharisma; Setianingsih, Susi; Hidayat, Galih B.; Ahmad, Tuhfa R.
Jurnal Teknik Informatika (Jutif) Vol. 5 No. 4 (2024): JUTIF Volume 5, Number 4, August 2024 - SENIKO
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2024.5.4.2181

Abstract

Cigarettes are an addictive substance that kills around 8 million people every year, as of 2022 there will be around 8,67 million deaths in the world caused by cigarettes and other tobacco products with resulting economic losses of around 2 trillion USD. Efforts to reduce losses due to smoking in Indonesia have been implemented through various regulations and rules that have been established, such as Law Number 36 of 2009 Article 115 concerning non-smoking areas. The target for non-smoking areas (NSA) regulations in Indonesia will reach 100% by 2023. However, currently, only 86% of regions have NSA regulations and must continue to monitor and evaluate through regulations set by the government. One solution to emphasize non-smoking areas with the latest technology connections to support Smart City is a smoke detection system using IoT. DesMoCam (Detection Smoking Camera) applies the latest machine learning model, InceptionResNet2, which has high accuracy and has the ability to detect smokers precisely in a Non-Smoking Area (NSA). DesMoCam uses a Raspberry Pi with ESP32-CAM to capture situations in a smoking-free room and warnings through the speaker. Machine learning modeling includes data acquisition with smoking and non-smoking images, data preprocessing, two-way modeling with and without a freeze layer, and analysis of model results. The InceptionResnet2 model used for image identification and classification, achieved an accuracy of 92.75%.
Co-Authors Ahmad, Tuhfa R. Hidayat, Galih B. Nayu, Balqist Kharisma Setianingsih, Susi