Lazaro, Alvin (2017) Deteksi Jenis Kendaraan di Jalan Menggunakan OpenCV. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
5113100067-Undergraduate Theses.pdf - Published Version Download (1MB) | Preview |
Abstract
Saat ini, jalan raya merupakan komponen penting dalam lalu lintas. Dalam setiap kota, tidak semua jalan dapat dilalui oleh semua jenis kendaraan. Untuk membantu memantau kondisi jalan, diperlukan suatu program yang bisa mendeteksi jenis kendaraan yang melalui jalan tersebut. Sebelumnya, sudah ada program-program serupa yang mampu mendeteksi kendaraan di jalan raya. Namun, program-program tersebut hanya melakukan deteksi kendaraan saja. Program-program tersebut tidak mampu melakukan deteksi terhadap jenis kendaraan di jalan raya. Untuk itu, Tugas Akhir ini adalah pengembangan lebih lanjut dari deteksi kendaraan yaitu melakukan deteksi jenis kendaraan serta menghitung kendaraan yang terdeteksi berdasarkan jenisnya. Tujuan Tugas Akhir ini yaitu untuk membuat program yang mampu mengidentifikasi jenis kendaraan pada suatu input video dan menghitung jumlah kendaraan yang terdeteksi berdasarkan jenisnya. Pengguna cukup memasukkan video rekaman lalu lintas ke program CarDetection. Nantinya, program akan memproses video yang dimasukkan dan menghasilkan file .txt sebagai keluaran. File ini berisi jenis kendaraan yang terdeteksi beserta jumlah kendaraan yang terdeteksi berdasarkan jenisnya. Dari hasil pengujian, program memiliki akurasi rata-rata 77.8% untuk kondisi jalan sepi, 47.5% untuk kondisi jalan normal, dan 28.2% untuk kondisi jalan padat. Hal-hal yang mempengaruhi akurasi deteksi adalah sudut pandang kendaraan terhadap kamera, jarak antar kendaraan, serta waktu tempuh kendaraan dalam memasuki hingga keluar dari wilayah deteksi.
============================================================================
Today, highways are an important component of traffic. In every city, not all roads can be passed by all types of vehicles. In order to help monitoring road conditions, a program that can detect the type of vehicle that passes through the road is needed. Previously, there are similar programs which can detect the vehicle on the road. However, these programs only do vehicle detection alone. These programs are unable to detect the type of vehicle on the highway. Therefore, this Final Project is the further development of vehicle detection, which is to detect the type of vehicle on the highway and calculate the number of vehicles detected based on its type. The purpose of this Final Project is to create a program which is able to identify the type of vehicle on a video input and calculate the number of vehicles detected based on its type. Users simply insert a video recording of traffic to the CarDetection program. Later, the program will process the inserted video and generate a .txt file as output. This file contains the type of vehicle detected along with the number of vehicles detected by type. From the test results, the program has an average accuracy of 77.8% for deserted road conditions, 47.5% for normal road conditions, and 28.2% for solid road conditions. Things that affect detection accuracy are the vehicle's viewing angle to the camera, the distance between vehicles, and the travel time of the vehicle in entering and exiting the detection area.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSIf 006.37 Laz d-1 |
| Uncontrolled Keywords: | Cascade, Classifier, Jenis kendaraan, Haar-like feature, OpenCV |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms |
| Divisions: | Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | ALVIN LAZARO |
| Date Deposited: | 31 Oct 2017 03:32 |
| Last Modified: | 05 Mar 2019 03:05 |
| URI: | http://repository.its.ac.id/id/eprint/42236 |
Actions (login required)
 |
View Item |