Searching near-replicas of images via clustering

Edward Y. Chang, Chen Li, James Wang, Peter Mork, Gio Wiederhold

Research output: Contribution to journalConference article

28 Citations (Scopus)

Abstract

Internet piracy has been one of the major concerns for Web publishing. In this study we present a system, RIME, that we have prototyped for detecting unauthorized image copying on the World-Wide Web. To speed up the copy detection, RIME uses a new clustering/hashing approach that first clusters similar images on adjacent disk cylinders and then builds indexes to access the clusters made in this way. Searching for the replicas of all image often takes just one IO to look up the location of the cluster containing similar objects and one sequential file IO to read in this cluster. Our experimental results show that RIME can detect image copies both more efficiently and effectively than the traditional content-based image retrieval systems that use tree-like structures to index images. In addition, RIME copes well with image format conversion, resampling, requantization and geometric transformations.

Original languageEnglish (US)
Pages (from-to)281-292
Number of pages12
JournalProceedings of SPIE - The International Society for Optical Engineering
Volume3846
StatePublished - Dec 1 1999

Fingerprint

Image Clustering
Copying
Image retrieval
Replica
replicas
World Wide Web
Internet
Geometric transformation
Content-based Image Retrieval
Hashing
Resampling
files
Speedup
Adjacent
format
retrieval
Clustering
Experimental Results

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Chang, Edward Y. ; Li, Chen ; Wang, James ; Mork, Peter ; Wiederhold, Gio. / Searching near-replicas of images via clustering. In: Proceedings of SPIE - The International Society for Optical Engineering. 1999 ; Vol. 3846. pp. 281-292.
@article{23c87e00512f4e9aa80b44382091c000,
title = "Searching near-replicas of images via clustering",
abstract = "Internet piracy has been one of the major concerns for Web publishing. In this study we present a system, RIME, that we have prototyped for detecting unauthorized image copying on the World-Wide Web. To speed up the copy detection, RIME uses a new clustering/hashing approach that first clusters similar images on adjacent disk cylinders and then builds indexes to access the clusters made in this way. Searching for the replicas of all image often takes just one IO to look up the location of the cluster containing similar objects and one sequential file IO to read in this cluster. Our experimental results show that RIME can detect image copies both more efficiently and effectively than the traditional content-based image retrieval systems that use tree-like structures to index images. In addition, RIME copes well with image format conversion, resampling, requantization and geometric transformations.",
author = "Chang, {Edward Y.} and Chen Li and James Wang and Peter Mork and Gio Wiederhold",
year = "1999",
month = "12",
day = "1",
language = "English (US)",
volume = "3846",
pages = "281--292",
journal = "Proceedings of SPIE - The International Society for Optical Engineering",
issn = "0277-786X",
publisher = "SPIE",

}

Searching near-replicas of images via clustering. / Chang, Edward Y.; Li, Chen; Wang, James; Mork, Peter; Wiederhold, Gio.

In: Proceedings of SPIE - The International Society for Optical Engineering, Vol. 3846, 01.12.1999, p. 281-292.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Searching near-replicas of images via clustering

AU - Chang, Edward Y.

AU - Li, Chen

AU - Wang, James

AU - Mork, Peter

AU - Wiederhold, Gio

PY - 1999/12/1

Y1 - 1999/12/1

N2 - Internet piracy has been one of the major concerns for Web publishing. In this study we present a system, RIME, that we have prototyped for detecting unauthorized image copying on the World-Wide Web. To speed up the copy detection, RIME uses a new clustering/hashing approach that first clusters similar images on adjacent disk cylinders and then builds indexes to access the clusters made in this way. Searching for the replicas of all image often takes just one IO to look up the location of the cluster containing similar objects and one sequential file IO to read in this cluster. Our experimental results show that RIME can detect image copies both more efficiently and effectively than the traditional content-based image retrieval systems that use tree-like structures to index images. In addition, RIME copes well with image format conversion, resampling, requantization and geometric transformations.

AB - Internet piracy has been one of the major concerns for Web publishing. In this study we present a system, RIME, that we have prototyped for detecting unauthorized image copying on the World-Wide Web. To speed up the copy detection, RIME uses a new clustering/hashing approach that first clusters similar images on adjacent disk cylinders and then builds indexes to access the clusters made in this way. Searching for the replicas of all image often takes just one IO to look up the location of the cluster containing similar objects and one sequential file IO to read in this cluster. Our experimental results show that RIME can detect image copies both more efficiently and effectively than the traditional content-based image retrieval systems that use tree-like structures to index images. In addition, RIME copes well with image format conversion, resampling, requantization and geometric transformations.

UR - http://www.scopus.com/inward/record.url?scp=0033340192&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033340192&partnerID=8YFLogxK

M3 - Conference article

VL - 3846

SP - 281

EP - 292

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

SN - 0277-786X

ER -