DETEKSI PLAT NOMOR KENDARAAN BERGERAK BERBASIS METODE YOU ONLY LOOK ONCE (YOLO)

Pratama, Ario Fajar (2020) DETEKSI PLAT NOMOR KENDARAAN BERGERAK BERBASIS METODE YOU ONLY LOOK ONCE (YOLO). Undergraduate thesis, Institut Teknologi Sepuluh Nopember Surabaya.

[img] Text
2020-06111640000087-buku.pdf - Accepted Version
Restricted to Repository staff only

Download (5MB) | Request a copy

Abstract

Semakin berkembangnya alat transportasi yang digunakan manusia, dibutuhkan sistem pengawasan lalu lintas yang canggih yang disebut dengan Intelligent Transportation System (ITS). Salah satu bagian dari ITS adalah sistem pendeteksian plat nomor kendaraan. Banyak penelitian sebelumnya berbasis deep learning telah dilakukan dalam pendeteksian objek benda, salah satunya adalah You Only Look Once (YOLO). Oleh sebab itu, pada penelitian ini dilakukan deteksi plat nomor kendaraan bergerak berbasis metode You Only Look Once (YOLO). Alur pendeteksian plat nomor ini diantaranya input video, akuisisi, input ROI, proses YOLO, tracking, memilih con�fidence tertinggi dan cropping. Berdasarkan penelitian ini didapatkan Mean Average Precision (mAP%) 87.89% dan hasil deteksi lokasi plat nomor yang terlihat secara empiris memiliki rata-rata akurasi sebesar 85.81% ================================================================================================================== The more developed means of transportation used by humans, we need a sophisticated tra�c control system called Intelligent Transportation System (ITS). One part of ITS is the vehicle number plate detection system. Many previous studies based on deep learning have been conducted in object detection, one of which is You Only Look Once (YOLO). Therefore, in this study, the detection of vehicle number plates based on the You Only Look Once (YOLO) method. The flow detection of this number is published video input, obtained, input ROI, YOLO process, tracking, select the highest confi�dence and cropping. Based on this research, it was obtained Mean Average Precision (mAP%) 87.89% and the results of detection license plates seen by empiricists had an average evaluation of 85.81%

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: YOLOV3, Deep Learning, Pengolahan Citra, Tracking ================================================================================================================== YOLOV3, Deep Learning, Image Processing, Tracking
Subjects: Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QA Mathematics > QA76.9.U83 Graphical user interfaces
T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User: Ario Fajar Pratama
Date Deposited: 25 Aug 2020 06:02
Last Modified: 25 Aug 2020 06:02
URI: http://repository.its.ac.id/id/eprint/79534

Actions (login required)

View Item View Item