Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Akbar Wira Bramantya
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer

 1 
Rancang Bangun Sistem Klasifikasi Tipe Permukaan Jalan menggunakan Gray Level Co-Occurrence Matrix (GLCM) dan Support Vector Machine (SVM) berbasis Raspberry Pi Akbar Wira Bramantya; Fitri Utaminingrum
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 2 (2022): Februari 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Roads are a pathways made to facilitate traffic access from one place to another. Indonesia has a roads as far as 537.838 km. However, all of these road didn't have same condition, there are still many road in bad condition. On the other side, our current transportation is just evolve to the Smart Car era, where this technology focuses on autonomous driving systems and driver safety. Therefore, needed a system that can overcome these various road conditions. So, in this research will do a testing using Gray Level Co-Occurrence Matrix (GLCM) and Support Vector Machine (SVM) to classify the roads are in asphalt condition, rocky condition, or paved condition. The GLCM feature used in this research is the best 4 features from total 6 features to be tested namely dissimilarity, correlation, homogeneity, contrast, ASM, and energy. This test is started using SVM linear kernel, polynomial kernel, and RBF kernel, the value of d = 1, and value of θ = 0o, 45o, 90o, 135o. After all of these tests, the best result were obtained. They are GLCM using dissimilarity, correlation, contrast, and energy features, the value of d = 1, and combination from all of angles, angle 0o, 45o, 90o, 135o simultaneously also for SVM using linear kernel. The accuracy obtained from this combination reaches 97% in SVM training set and 98,3% in manual testing using test data. System integration testing using camera input got 88,6% in accuracy and system integration testing on electric motors performance got 87% in accuracy.
Co-Authors Utaminingrum, Fitri