RANCANG BANGUN SISTEM DETEKSI DINI KEBAKARAN BERBASIS ESP32-CAM DAN SENSOR ASAP MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK UNTUK KLASIFIKASI VISUAL API

Ajie Alvido Akraam, . (2026) RANCANG BANGUN SISTEM DETEKSI DINI KEBAKARAN BERBASIS ESP32-CAM DAN SENSOR ASAP MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK UNTUK KLASIFIKASI VISUAL API. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

[img] Text
ABSTRAK.pdf
Download (241kB)
[img] Text
AWAL.pdf
Download (1MB)
[img] Text
BAB 1.pdf
Restricted to Repository UPNVJ Only
Download (352kB)
[img] Text
BAB 2.pdf
Restricted to Repository UPNVJ Only
Download (456kB)
[img] Text
BAB 3.pdf
Restricted to Repository UPNVJ Only
Download (642kB)
[img] Text
BAB 4.pdf
Restricted to Repository UPNVJ Only
Download (848kB)
[img] Text
BAB 5.pdf
Download (288kB)
[img] Text
DAFTAR PUSTAKA.pdf
Download (251kB)
[img] Text
RIWAYAT HIDUP.pdf
Restricted to Repository UPNVJ Only
Download (176kB)
[img] Text
LAMPIRAN.pdf
Restricted to Repository UPNVJ Only
Download (5MB)
[img] Text
HASIL PLAGIARISME.pdf
Restricted to Repository staff only
Download (19MB)
[img] Text
ARTIKEL KI.pdf
Restricted to Repository staff only
Download (578kB)

Abstract

Conventional fire detection systems often trigger false alarms as they rely solely on smoke or temperature sensors without visual verification, thereby reducing trust and emergency response effectiveness. This study designs and builds an Internet of Things (IoT)-based early fire detection system integrating an ESP32-CAM microcontroller and an MQ-2 smoke sensor, utilizing the Convolutional Neural Network (CNN) method with the YOLO algorithm for real-time visual fire classification. The system operates with a dual verification mechanism where warning notifications and visual evidence are sent via Telegram, and sprinkler actuators are automatically activated only when both fire and smoke indications are detected simultaneously. Test results demonstrate that the system detects fire with an accuracy rate above 90% and an average response time between 2.1 to 4.8 seconds, making it an adaptive and reliable solution for enhancing safety in enclosed spaces.

Item Type: Thesis (Skripsi)
Additional Information: No. Panggil:2110314015 [Pembimbing 1: Achmad Zuchriadi} {Pembimbing 2: Ni Putu Devira Ayu Martini} {P:enguji 1:Muhamad Alif Razi} {Penguji 2: Andre Suwardana Adiwidya}
Uncontrolled Keywords: Early Fire Detection, ESP32-CAM, MQ-2 Sensor, Convolutional Neural Network (CNN), Internet of Things (IoT).
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Fakultas Teknik > Program Studi Teknik Elektro (S1)
Depositing User: AJIE ALVIDO AKRAAM
Date Deposited: 12 Feb 2026 08:38
Last Modified: 12 Feb 2026 08:38
URI: http://repository.upnvj.ac.id/id/eprint/42494

Actions (login required)

View Item View Item