MSAI: Masking Sensitive Area of Image on IoT Cameras

Liu, J; Liu, Y; Cui, L; Yu, S

MSAI: Masking Sensitive Area of Image on IoT Cameras

Liu, J Liu, Y Cui, L Yu, S

Permalink

Publisher:: LIBRARY & INFORMATION CENTER, NAT DONG HWA UNIV
Publication Type:: Journal Article
Citation:: Journal of Internet Technology, 2021, 22, (7), pp. 1553-1561
Issue Date:: 2021-01-01

Recently Added

	Filename	Description	Size
	2626-3566-1-SM.pdf	Published version	3.94 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is new to OPUS and is not currently available.

Full metadata record

Field	Value	Language
dc.contributor.author	Liu, J
dc.contributor.author	Liu, Y
dc.contributor.author	Cui, L
dc.contributor.author	Yu, S https://orcid.org/0000-0003-4485-6743
dc.date.accessioned	2022-09-21T02:49:19Z
dc.date.available	2022-09-21T02:49:19Z
dc.date.issued	2021-01-01
dc.identifier.citation	Journal of Internet Technology, 2021, 22, (7), pp. 1553-1561
dc.identifier.issn	1607-9264
dc.identifier.issn	2079-4029
dc.identifier.uri	http://hdl.handle.net/10453/161934
dc.description.abstract	In smart cities, images captured by Internet of Things (IoT) cameras are transmitted to data center via intermediate nodes. The shared images can reveal much of sensitive information of users, which has caused increasing privacy concerns. Various encryption-based techniques have been developed for privacy preserving. However, they are not suitable for IoT cameras due to the high computation cost. In this paper, we propose a lightweight image privacy protection scheme considering personalized protection requirements of users. We first introduce an object detection algorithm to detect and pick up sensitive areas in images according to the specific requirements of users. Then, we propose a membranebased method to mask the sensitive areas before uploading images to the data center. In particular, the masking operation does not require much computing resources on the used cloud platform, and the masking size of membrane can be dynamically adjusted. Our experiments on real-world datasets demonstrate the effectiveness and feasibility of the proposed scheme.
dc.language	English
dc.publisher	LIBRARY & INFORMATION CENTER, NAT DONG HWA UNIV
dc.relation.ispartof	Journal of Internet Technology
dc.relation.isbasedon	10.53106/160792642021122207009
dc.rights	info:eu-repo/semantics/restrictedAccess
dc.subject	0805 Distributed Computing, 0806 Information Systems, 1005 Communications Technologies
dc.subject.classification	Networking & Telecommunications
dc.title	MSAI: Masking Sensitive Area of Image on IoT Cameras
dc.type	Journal Article
utslib.citation.volume	22
utslib.for	0805 Distributed Computing
utslib.for	0806 Information Systems
utslib.for	1005 Communications Technologies
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	recently_added	*
dc.date.updated	2022-09-21T02:49:13Z
pubs.issue	7
pubs.publication-status	Published
pubs.volume	22
utslib.citation.issue	7

Abstract:

In smart cities, images captured by Internet of Things (IoT) cameras are transmitted to data center via intermediate nodes. The shared images can reveal much of sensitive information of users, which has caused increasing privacy concerns. Various encryption-based techniques have been developed for privacy preserving. However, they are not suitable for IoT cameras due to the high computation cost. In this paper, we propose a lightweight image privacy protection scheme considering personalized protection requirements of users. We first introduce an object detection algorithm to detect and pick up sensitive areas in images according to the specific requirements of users. Then, we propose a membranebased method to mask the sensitive areas before uploading images to the data center. In particular, the masking operation does not require much computing resources on the used cloud platform, and the masking size of membrane can be dynamically adjusted. Our experiments on real-world datasets demonstrate the effectiveness and feasibility of the proposed scheme.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/161934