Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/163997| Title: | Implementasi CCTV Berbasis Kecerdasan Buatan untuk Pemantauan Alat Pelindung Diri pada Buruh Proyek Menggunakan Algoritma YOLOv8 |
| Other Titles: | Implementation of Artificial Intelligence Based CCTV for Monitoring Personal Protective Equipment for Project Workers Using the YOLOv8 Algorithm |
| Authors: | Fathonah, Lathifunnisa Dani, Mohamad Fikih Amar |
| Issue Date: | 2025 |
| Publisher: | IPB University |
| Abstract: | Tingginya angka kecelakaan kerja di sektor konstruksi sering kali berkaitan dengan rendahnya kepatuhan penggunaan Alat Pelindung Diri (APD), sementara pengawasan manual oleh petugas Kesehatan dan Keselamatan Kerja (K3) bersifat terbatas. Penelitian ini bertujuan untuk mengembangkan sistem pemantauan otomatis berbasis kecerdasan buatan (Artificial Intelligence, AI) untuk meningkatkan standar keselamatan kerja. Metode yang digunakan adalah implementasi sistem Closed-Circuit Television (CCTV) yang terintegrasi dengan perangkat Raspberry Pi 5 dan akselerator AI Hailo-8L. Sistem ini menjalankan model deteksi objek YOLOv8 untuk mengidentifikasi penggunaan APD seperti helm, rompi, dan sepatu keselamatan secara real-time. Hasil penelitian menunjukkan bahwa sistem mampu mendeteksi pelanggaran secara otomatis dan mengirimkan datanya ke sebuah dasbor web pemantauan. Dengan implementasi sistem ini, diharapkan pengawasan terhadap kepatuhan penggunaan APD dapat menjadi lebih ketat dan respons terhadap pelanggaran menjadi lebih cepat, sehingga berpotensi mengurangi risiko kecelakaan kerja di lingkungan proyek. The high rate of work-related accidents in the construction sector is often associated with low compliance in the use of Personal Protective Equipment (PPE), while manual supervision by Health, Safety, and Environment (HSE) officers is limited. This research aims to develop an automated monitoring system based on artificial intelligence (AI) to improve work safety standards. The method used is the implementation of a Closed-Circuit Television (CCTV) system integrated with a Raspberry Pi 5 device and a Hailo-8L AI accelerator. This system runs the YOLOv8 object detection model to identify the use of PPE such as helmets, vests, and safety boots in real-time. The results indicate that the system can automatically detect violations and send the data to a web-based monitoring dashboard. With the implementation of this system, it is expected that supervision of PPE compliance can become more stringent and responses to violations can be faster, thereby potentially reducing the risk of work accidents in project environments. |
| URI: | http://repository.ipb.ac.id/handle/123456789/163997 |
| Appears in Collections: | UT - Computer Engineering Tehcnology |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_J0304211089_ea279c890cfa4ff1acde1b138d83e97e.pdf | Cover | 1.08 MB | Adobe PDF | View/Open |
| fulltext_J0304211089_4a0bb26fdcda4260b6c250ae716e9330.pdf Restricted Access | Fulltext | 3.33 MB | Adobe PDF | View/Open |
| lampiran_J0304211089_e7683da56b2e423e9ef763fcf1e76838.pdf Restricted Access | Lampiran | 470.32 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.