eprintid: 3850
rev_number: 10
eprint_status: archive
userid: 285
dir: disk0/00/00/38/50
datestamp: 2019-12-19 03:17:26
lastmod: 2019-12-19 03:17:26
status_changed: 2019-12-19 03:17:26
type: conference_item
metadata_visibility: show
creators_name: Zhu, Qiuchen
creators_name: Phung, Manh Duong
creators_name: Quang, Ha
creators_id: duongpm@vnu.edu.vn
creators_id: quang.ha@uts.edu.au
title: Crack Detection Using Enhanced Hierarchical Convolutional Neural Networks
ispublished: pub
subjects: ECE
subjects: ElectronicsandComputerEngineering
divisions: fac_fet
abstract: Unmanned aerial vehicles (UAV) are expected to replace human in hazardous tasks of surface inspection due to their flexibility in operating space and capability of collecting high quality visual data. In this study, we propose enhanced hierarchical convolutional neural networks (HCNN) to detect cracks from image data collected by UAVs. Unlike traditional HCNN, here a set of branch networks is utilised to reduce the obscuration in the down-sampling process. Moreover, the feature preserving blocks combine the current and previous terms from the convolutional blocks to provide input to the loss functions in order to reduce the weight of resized images and thus minimise the information loss. Experiments on images of different crack datasets have been carried out to demonstrate the effectiveness of the proposed HCNN.
date: 2019-12-11
date_type: published
official_url: https://ssl.linklings.net/conferences/acra/acra2019_proceedings/views/includes/files/pap104s1-file1.pdf
full_text_status: public
pres_type: paper
event_title: Australasian Conference on Robotics and Automation 2019
event_location: Adelaide, Australia
event_dates: 9-11 December 2019
event_type: conference
refereed: TRUE
citation:   Zhu, Qiuchen and Phung, Manh Duong and Quang, Ha  (2019) Crack Detection Using Enhanced Hierarchical Convolutional Neural Networks.  In: Australasian Conference on Robotics and Automation 2019, 9-11 December 2019, Adelaide, Australia.     
document_url: https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3850/1/pap104s1-file1.pdf