CD Skripsi

Sistem Pendeteksi Pengguna Helm Di Dalam Atm Dengan Metode Deep Learning Berbasis Raspberry Pi 4

---

ABSTRACT
Crime is all kinds of activities carried out by the whole community because it violates the law, social and religion and harms many parties. One of the crimes that often occurs is a crime related to an ATM (Automatic Teller Machine). Perpetrators who enter ATMs usually carry out their actions using head coverings such as helmets, hats, skullcaps and others in committing crimes. To minimize the occurrence of criminal acts that occur at ATMs, this study aims to detect head coverings that are worn when entering an ATM. This object detection system uses a Raspberry Pi 4 mini computer board as a controller, a Pi camera as a sensor to detect objects, speakers to provide information in the form of alarms and telegram bots to receive image messages with detected objects. The method used in this study is YOLO (You Only Look Once) which is an algorithm used for direct object detection. YOLO works by looking at the entire image once, then the neural network immediately detects existing objects. This system consists of three main processes, namely the pre-processing process, the training process and the detection process. The pre-processing process is resizing and labeling annotations on the image dataset, then the training process on the dataset using YOLOv3. In the detection process, this system performs localization and classification with a single step process so that the result of this detection process is a person wearing a helmet. From the results of testing this system it was stated that the system was able to detect with a fairly high accuracy of 96%.
Keywords: Crime, Headgear, ATM (Automated Teller Machine), Raspberry Pi 4, YOLO (You Only Look Once).

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

1807111363

Penerbit

Pekanbaru : Universitas Riau – Fakultas Teknik – Teknik Informatika., 2023

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

1807111363

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

TEKNIK ELEKTRO

Info Detail Spesifik

-

Pernyataan Tanggungjawab

DAUS

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar