Susilo, Agus and Hendriyawan A, M.S (2019) RANCANG BANGUN SISTEM PENDETEKSI PELANGGARAN PADA JALUR BUSWAY BERBASIS CITRA DIGITAL MENGGUNAKAN METODE SSD (SINGLE SHOT MULTIBOX DETECTOR). Tugas Akhir thesis, University of Technology Yogyakarta.
|
Text
NASKAH PUBLIKASI_TA_AGUS SUSILO_5150711022.pdf Download (1MB) | Preview |
Abstract
ABSTRACT In this age there are still many traffic violations that occur anywhere. Many violations were committed by motorists who did not pay attention to traffic signs and entered the busway lane specifically for Transjakarta buses. The problem faced in this study is how the system can distinguish between Tranjakarta buses with other vehicles, namely digital image-based cars and motorcycles. This study uses the SSD method (Single Shot MultiBox Detector) for the process of training datasets and running systems. The components used in this study are laptops, python IDLE as an application for coding entire studies, tensorflow libraries that have deep learning as a feature extraction process, and webcam camera to detect vehicle objects. Then the four scenarios were tested directly on the prototype. If a vehicle in the form of a car and motorbike was detected on the busway lane, a violation occurred. When a Transjakarta bus drove on the busway lane, there was no detection of a violation or safe. The output of this detection is a notification of a text that says "Violation" and "Safe" on the monitor's LCD laptop display. The results of this study carried out a test of the accuracy of vehicle object detection from an existing dataset in the form of photos of buses, cars, and motorcycles which obtained an average percentage of accuracy of 98.2%. so that the accuracy of the vehicle datasets reaches the target of > 70%. Key words : Digital Image, Dataset, Deep Learning, Tensorflow, SSD (Single Shot MultiBox Detector)
| 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: | 02 Aug 2019 04:02 |
| Last Modified: | 02 Aug 2019 04:02 |
| URI: | http://eprints.uty.ac.id/id/eprint/3330 |
Actions (login required)
 |
View Item |
