FATURACHMAN, FAJAR (2023) IMPLEMENTASI DEEP LEARNING UNTUK DETEKSI ALAT PELINDUNG DIRI DENGAN MODEL ALGORITMA YOLOV5. S2 thesis, Universitas Mercu Buana - Menteng.
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (Cover)
55420120012-FAJAR FATURACHMAN-01COVER - fajar faturachman.pdf Download (3MB) |
|
Text (Abstrak)
55420120012-FAJAR FATURACHMAN-02 ABSTRAK - fajar faturachman.pdf Download (670kB) |
|
Text (Bab 1)
55420120012-FAJAR FATURACHMAN-03 BAB 1 - fajar faturachman.pdf Restricted to Registered users only Download (1MB) |
|
Text (Bab 2)
55420120012-FAJAR FATURACHMAN-04 BAB 2 - fajar faturachman.pdf Restricted to Registered users only Download (3MB) |
|
Text (Bab 3)
55420120012-FAJAR FATURACHMAN-05 BAB 3 - fajar faturachman.pdf Restricted to Registered users only Download (3MB) |
|
Text (Bab 4)
55420120012-FAJAR FATURACHMAN- 06 BAB 4 - fajar faturachman.pdf Restricted to Registered users only Download (9MB) |
|
Text (Bab 5)
55420120012-FAJARFATURACHMAN-07 BAB 5 - fajar faturachman.pdf Restricted to Registered users only Download (711kB) |
|
Text (Daftar Pustaka)
55420120012-FAJAR FATURACHMAN-09 DAFTAR PUSTAKA - fajar faturachman.pdf Restricted to Registered users only Download (768kB) |
|
Text (Lampiran)
55420120012-FAJAR FATURACHMAN-10 LAMPIRAN - fajar faturachman.pdf Restricted to Registered users only Download (4MB) |
|
Text (Form Keabsahan)
55420120012-FAJAR FATURACHMAN-11 HASIL SCAN FORMULIR PERNYATAAN KEABSAHAN DAN PERSETUJUAN PUBLIKASI TUGAS AKHIR - fajar faturachman.pdf Restricted to Registered users only Download (1MB) |
Abstract
Penggunaan alat pelindung diri atau APD merupakan kewajiban bagi para pekerja di dunia industri kontruksi, sesuai dengan Peraturan Menteri Tenaga Kerja dan Transmigrasi Republik Indonesia Nomor Per.08/MEN/VII/2010 tentang pengunaan alat pelindung diri. Penerapan teknologi kecerdasan buatan dan pembelajaran mesin untuk keamanan bangunan semakin meningkat. Perkembagan teknologi dalam dunia deteksi objek sudah sampai tahap menyamai kinerja alat panca indera manusia, You Only Look Once atau disebut Yolo secara filosofi bisa mendeteksi objek dnegan sekali lihat dengan model Convolutional Neural Network dengan metode yang efektif dan efisien, Yolov5 merupakan versi ke 5 dari Yolo yang memiliki arsitektur yang mampu menghasilkan akurasi yang lebih baik dari versi sebelumnya. Sehingga metode yang digunakan dalam penelitian ini adalah YOLO (You Only Look Once) versi 5. Penelitan ini akan membahas tentang implementasi Model Algoritma YOLOv5 untuk mendeteksi Alat pelindung Diri, Objek yang akan di deteksi adalah perangkat Alat Pelindung Diri, yaitu Helm Pengaman, Kacamata Pengaman, Masker dan Penutup Telinga. Hasil Penelitian dapat disimpulkan nilai akurasi saat pengujian mendapatkan nilai yang cukup baik yaitu 94% pada deteksi Alat Pelindung Diri dengan Model YOLOv5x dengan Batch 160 dan Epoch Sebanyak 300. The use of personal protective equipment or PPE is an obligation for workers in the construction industry, in accordance with the Regulation of the Minister of Manpower and Transmigration of the Republic of Indonesia Number Per.08/MEN/VII/2010 concerning the use of personal protective equipment. The application of artificial intelligence and machine learning technologies to building security is increasing. Technological developments in the world of object detection have reached the stage of matching the performance of human sensory organs. You Only Look Once or called Yolo, philosophically, can detect objects with one look using the Convolutional Neural Network model with an effective and efficient method. Yolov5 is the 5th version of Yolo which has an architecture capable of producing better accuracy than the previous version. So that the method used in this study is YOLO (You Only Look Once) version 5. This research will discuss the implementation of the YOLOv5 Algorithm Model to detect Personal Protective Equipment, the object to be detected is a Personal Protective Equipment device in accordance with the Minister of Manpower Regulations. and Transmigration of the Republic of Indonesia, namely Safety Helmets, Safety Goggles, Masks and Earmuffs. The results of the study can be concluded that the accuracy value when testing gets a fairly good value, namely 94% in the detection of Personal Protective Equipment with the YOLOv5x Model with a Batch of 160 and an Epoch of 300. Keywords: Digital Image Processing, Personal Protective Equipment, YOLOv5
| Item Type: | Thesis (S2) |
|---|---|
| NIM/NIDN Creators: | 55420120012 |
| Uncontrolled Keywords: | Digital Image Processing, Personal Protective Equipment, YOLOv5 |
| Subjects: | 600 Technology/Teknologi > 620 Engineering and Applied Operations/Ilmu Teknik dan operasi Terapan > 621 Applied Physics/Fisika terapan |
| Divisions: | Pascasarjana > Magister Teknik Elektro |
| Depositing User: | MELATI CAHYA FITRIANI |
| Date Deposited: | 24 Mar 2023 07:16 |
| Last Modified: | 24 Mar 2023 07:16 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/75424 |
Actions (login required)
 |
View Item |
