Abstract

Damage roads had been disturbed by social activities and involved in traffic accidents. Identification and classification of the types of defected road are required to minimize its impact and before repairs. Digital image processing technology can identify and classify the type of damaged roads automatically. In this study, the classification of defected roads is automatic with a multiclass Support Vector Machine(SVM). There are three classes in the classification process, namely, alligators, potholes, and cracks. The process of recognizing defected roads uses a multiclass SVM classification model with polynomial and Gaussian kernel function and One Vs. All strategy and uses a cell size of 16 × 16 pixels during the Histogram of Oriented Gradients (HOG) feature extraction process. and produces an accuracy value of 78,85%.

Original languageEnglish
Article number012048
JournalJournal of Physics: Conference Series
Volume1821
Issue number1
DOIs
Publication statusPublished - 29 Mar 2021
Event6th International Conference on Mathematics: Pure, Applied and Computation, ICOMPAC 2020 - Surabaya, Virtual, Indonesia
Duration: 24 Oct 2020 → …

Fingerprint

Dive into the research topics of 'Classification of Damaged Road Types Using Multiclass Support Vector Machine (SVM)'. Together they form a unique fingerprint.

Cite this