Prasetyo, Riyan Budi (2021) RANCANG BANGUN SISTEM STARTER MOBIL MENGGUNAKAN PENGENALAN WAJAH DENGAN METODE DEEP LEARNING. Tugas Akhir thesis, University of Technology Yogyakarta.
|
Text
Abstrak_5160711029_Riyan Budi Prasetyo - Ryan budi Prasetyo done.pdf Download (16kB) | Preview |
Abstract
ABSTRACT Indonesia is one of the countries where the population of each household has a private four-wheeled vehicle, namely a car. In 2018 according to the Central Bureau of Statistics, four-wheeled vehicles have reached a very high number of 16,440,987 and not including goods transporting vehicles. Every factory that produces cars must have thought carefully about the car security system, including by providing anti-theft alarms on cars, and so far only alarms have been used as a tool for security on cars, especially on mediocre cars. This causes the level of security on four-wheeled vehicles is still very minimal and less effective if the security system only uses alarms. Therefore we need an additional tool that is used as a car safety system, thereby increasing the security system for four-wheeled vehicles. In this study, a car starter tool will be made using facial recognition with the Deep Learning method in the form of a prototype. This tool can distinguish the faces of car owners and non-car owners by classifying facial files that have been scanned and those that have been trained and stored in the previous dataset. If the results of the extraction and face classification are similar, the car can be started and vice versa if the scanned face is not in the dataset or not in the previous training, the car will not be able to be started. Face samples were obtained from the webcam camera already installed on the steering wheel and connected to the Raspberry Pi and the Raspberry Pi Microcontroller was used to process the face recognition system. Based on the tests carried out, the confidence level or system confidence value was 76.80222048362440 % for the trained sample and 48.06133011 % for the untrained sample, and the confusion matrix or face recognition accuracy rate was 95%. The reliability of the car starter system using facial recognition with the Deep Learning method is considered to be quite good and the system can detect and distinguish faces with a fairly high level of accuracy. Keywords: Security, car, face recognition, Deep Learning, Raspberry Pi, Webcam.
| Item Type: | Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir) |
|---|---|
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering |
| Divisions: | Fakultas Sains Dan Teknologi > S1 Teknik Elektro |
| Depositing User: | Kaprodi Teknik Elektro |
| Date Deposited: | 26 Jun 2021 03:05 |
| Last Modified: | 26 Jun 2021 03:05 |
| URI: | http://eprints.uty.ac.id/id/eprint/7638 |
Actions (login required)
 |
View Item |
