Autonomous road potholes detection on video

Jia Juang Koh, Timothy Tzen Vun Yap*, Hu Ng, Vik Tor Goh, Hau Lee Tong, Chiung Ching Ho, Thiam Yong Kuek

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

This research work explores the possibility of using deep learning to produce an autonomous system for detecting potholes on video to assist in road monitoring and maintenance. Video data of roads was collected using a GoPro camera mounted on a car. Region-based Fully Convolutional Networks (RFCN) was employed to produce the model to detect potholes from images, and validated on the collected videos. The R-FCN model is able to achieve a Mean Average Precision (MAP) of 89% and a True Positive Rate (TPR) of 89% with no false positive.

Original languageEnglish
Title of host publicationComputational Science and Technology
EditorsRayner Alfred, Ag Asri Ag Ibrahim, Yuto Lim, Patricia Anthony
PublisherSpringer
Pages137-143
Number of pages7
ISBN (Electronic)9789811326226
ISBN (Print)9789811326219
DOIs
Publication statusPublished - 2019
Event5th International Conference on Computational Science and Technology 2018 - Kota Kinabalu, Malaysia
Duration: 29 Aug 201830 Aug 2018

Publication series

NameLecture Notes in Electrical Engineering
Volume481
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

Conference5th International Conference on Computational Science and Technology 2018
Abbreviated titleICCST 2018
Country/TerritoryMalaysia
CityKota Kinabalu
Period29/08/1830/08/18

Keywords

  • Deep learning
  • Machine learning
  • Object identification
  • Road surface defects
  • Video data

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Fingerprint

Dive into the research topics of 'Autonomous road potholes detection on video'. Together they form a unique fingerprint.

Cite this