Fatmawati, Intan (2018) Deteksi Kendaraan Roda Empat Untuk Mendukung Keamanan Berkendara Menggunakan Histogram Of Oriented Gradients Dan Support Vector Machine Berbasis Raspberry Pi. Sarjana thesis, Universitas Brawijaya.
Abstract
Keselamatan berkendara merupakan hal terpenting dalam berkendara dijalan raya agar terhindar dari kecelakaan. Kecelakaan terjadi akibat beberapa faktor diantaranya tidak konsentrasi saat mengemudi, mengantuk, dan lain sebagainya. Salah satu upaya untuk menghindari atau mengurangi resiko kecelakaan pada penelitian ini yaitu menjaga jarak kendaraan dengan kendaraan di depannya menggunakan image processing. Penelitian ini Histogram of Oriented Gradients (HOG) digunakan untuk ekstraksi mobil, SVM digunakan untuk membedakan kelas mobil dan bukan mobil, dan juga menggunakan kamera Raspberry Pi yang dipasang pada dashboard guna mendeteksi kendaraan didepannya. Jika jarak kendaraan dengan kendaraan didepannya kurang dari sama dengan 15 meter (≤15m) maka buzzer akan berbunyi sebagai tanda bahwa jarak kendaraan terlalu dekat. Hasil akurasi sistem dalam mendeteksi mobil menggunakan Support Vector Machine (SVM) berdasarkan fitur Histogram of Oriented Gradients (HOG) dengan pengujian pada jarak 10 m, 15 m, 20 m, 30 m sebesar 81.3%, untuk pengujian akurasi integrasi Hardware dan Software sebesar 87.5%.
English Abstract
Driving safety is the most important thing in driving on the highway to avoid accidents. Accidents occur due to several factors including lack of concentration while driving, drowsiness, and so forth. One effort to avoid or reduce the risk of accidents when driving is to maintain the distance of the vehicle with the vehicle in front of it. In this study the theme of image processing is used to achieve the goal of maintaining distance between vehicle by utilizing the Histogram of Oriented Gradients (HOG) method as a way of extracting vehicle features which in this case are cars, then classified by the Vector Support Machine (SVM) method to distinguish between car classes and not cars, the two methods are implemented uses a Raspberry Pi camera mounted on the dashboard to detect the vehicle in front of it. When the distance of the vehicle with the vehicle in front is less than or equal to 15 meters (≤15m), the buzzer will sound as a sign that the vehicle is too close. The accuracy of the system in detecting cars using Support Vector Machine (SVM) that based on the Histogram of Oriented Gradients (HOG) feature by testing at a distance of 10 m, 15 m, 20 m, dan 30 m is 81.3% and the testing accuracy of Hardware and Software integration is 87.5%.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/FTIK/2018/960/051900708 |
| Uncontrolled Keywords: | kendaraan, histogram of oriented gradients, support vector machine / detecting vehicles, histogram of oriented gradients, support vector machine |
| Subjects: | 300 Social sciences > 388 Transportation > 388.04 Special topics of transportation |
| Divisions: | Fakultas Ilmu Komputer > Teknik Informatika |
| Depositing User: | Endang Susworini |
| Date Deposited: | 16 Apr 2020 14:34 |
| Last Modified: | 22 Jun 2022 07:30 |
| URI: | http://repository.ub.ac.id/id/eprint/167050 |
![[thumbnail of Intan Fatmawati.pdf]](http://repository.ub.ac.id/style/images/fileicons/text.png) |
Text
Intan Fatmawati.pdf Download (4MB) |
Actions (login required)
 |
View Item |