Real-time automated road damages inspection using deep convolutional neural networks

Shahrul Munir, Mohamad Faiq Mohd and Bustamam, Muhammad Amiruddin and Ismail, Amelia Ritahani and Md Yusof, Norlia and Amir Hussin, Amir 'Aatieff (2023) Real-time automated road damages inspection using deep convolutional neural networks. International Journal on Perceptive and Cognitive Computing (IJPCC), 9 (1). pp. 122-127. E-ISSN 2462-229X

PDF (Real-Time Automated Road Damages Inspection Using Deep Convolutional Neural Networks) - Published Version Restricted to Registered users only Download (303kB) | Request a copy

Abstract

This study focused on developing a real time automated road damage inspection using deep neural networks. The performance of the build image detection model is evaluated to get the best overall result. Thousands of images from selected dataset are trained using You Only Look Once (YOLO) v4 algorithm which based on Convolutional Neural Network (CNN). The model is deployed into smartphones to take advantage of its availability camera. The road damage inspection application (app) can help the road users and municipalities in inspecting the road surface. Thus, it can prevent heavy damages to a vehicle and help in providing a better road damage maintenance management.

Item Type:	Article (Journal)
Uncontrolled Keywords:	image detection, Convolutional Neural Network, deep learning, YOLO,
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology > Department of Computer Science Kulliyyah of Information and Communication Technology > Department of Computer Science
Depositing User:	Amelia Ritahani Ismail
Date Deposited:	02 Aug 2023 15:06
Last Modified:	24 Jan 2024 16:22
URI:	http://irep.iium.edu.my/id/eprint/105793

Actions (login required)

View Item