Skip to main navigation Skip to search Skip to main content

Comparison of Support Vector Machine Classifier and Naïve Bayes Classifier on Road Surface Type Classification

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This study describes the comparison of road surface classification results using Support Vector Machine (SVM) classification and Naive Bayes classification. The dataset in this study is a collection of sub-images (750 images) from the road area of the image from Google Street View. From the dataset, 600 images as data training and 150 images as data testing. Texture features are extracted from road surface images in the dataset and then we build SVM and Naive Bayes classifiers to classify road surface images in 3 categories, asphalt, gravel, and paving. Evaluation of performance classification using precision, recall, f-measure, and accuracy. The results show that SVM classifier accuracy better than Naive Bayes classifier.

Original languageEnglish
Title of host publication3rd International Conference on Sustainable Information Engineering and Technology, SIET 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages48-53
Number of pages6
ISBN (Electronic)9781538674079
DOIs
Publication statusPublished - 2 Jul 2018
Event3rd International Conference on Sustainable Information Engineering and Technology, SIET 2018 - Malang, Indonesia
Duration: 10 Nov 201812 Nov 2018

Publication series

Name3rd International Conference on Sustainable Information Engineering and Technology, SIET 2018 - Proceedings

Conference

Conference3rd International Conference on Sustainable Information Engineering and Technology, SIET 2018
Country/TerritoryIndonesia
CityMalang
Period10/11/1812/11/18

Keywords

  • GLCM
  • Naïve Bayes
  • SVM
  • Texture analysis

Fingerprint

Dive into the research topics of 'Comparison of Support Vector Machine Classifier and Naïve Bayes Classifier on Road Surface Type Classification'. Together they form a unique fingerprint.

Cite this