Wavelet-based image indexing techniques with partial sketch retrieval capability

James Wang, Gio Wiederhold, Oscar Firschein, Sha Xin Wei

Research output: Contribution to conferencePaper

73 Citations (Scopus)

Abstract

This paper describes WBIIS (Wavelet-Based Image Indexing and Searching), a new image indexing and retrieval algorithm with partial sketch image searching capability for large image databases. The algorithm characterizes the color variations over the spatial extent of the image in a manner that provides semantically-meaningful image comparisons. The indexing algorithm applies a Daubechies' wavelet transform for each of the three opponent color components. The wavelet coefficients in the lowest few frequency bands, and their variances, are stored as feature vectors. To speed up retrieval, a two-step procedure is used that first does a crude selection based on the variances, and then refines the search by performing a feature vector match between the selected images and the query. For better accuracy in searching, two level multiresolution matching may also be used. Masks are used for partial-sketch queries. This technique performs much better in capturing coherence of image, object granularity, local color/texture, and bias avoidance than traditional color layout algorithms. When tested on a database of more than 10,000 general-purpose images, WBIIS is much faster and more accurate than traditional algorithms.

Original languageEnglish (US)
Pages13-24
Number of pages12
StatePublished - Jan 1 1997
EventProceedings of the 1997 IEEE International Forum on Research and Technology Advances in Digital Libraries, ADL'97 - Washington, DC, USA
Duration: May 7 1997May 9 1997

Other

OtherProceedings of the 1997 IEEE International Forum on Research and Technology Advances in Digital Libraries, ADL'97
CityWashington, DC, USA
Period5/7/975/9/97

Fingerprint

Color
Wavelet transforms
Frequency bands
Masks
Textures
Wavelets
Indexing
Query
Data base

All Science Journal Classification (ASJC) codes

  • Management of Technology and Innovation

Cite this

Wang, J., Wiederhold, G., Firschein, O., & Wei, S. X. (1997). Wavelet-based image indexing techniques with partial sketch retrieval capability. 13-24. Paper presented at Proceedings of the 1997 IEEE International Forum on Research and Technology Advances in Digital Libraries, ADL'97, Washington, DC, USA, .
Wang, James ; Wiederhold, Gio ; Firschein, Oscar ; Wei, Sha Xin. / Wavelet-based image indexing techniques with partial sketch retrieval capability. Paper presented at Proceedings of the 1997 IEEE International Forum on Research and Technology Advances in Digital Libraries, ADL'97, Washington, DC, USA, .12 p.
@conference{b27b8faa77854ff78405a87ae02a1751,
title = "Wavelet-based image indexing techniques with partial sketch retrieval capability",
abstract = "This paper describes WBIIS (Wavelet-Based Image Indexing and Searching), a new image indexing and retrieval algorithm with partial sketch image searching capability for large image databases. The algorithm characterizes the color variations over the spatial extent of the image in a manner that provides semantically-meaningful image comparisons. The indexing algorithm applies a Daubechies' wavelet transform for each of the three opponent color components. The wavelet coefficients in the lowest few frequency bands, and their variances, are stored as feature vectors. To speed up retrieval, a two-step procedure is used that first does a crude selection based on the variances, and then refines the search by performing a feature vector match between the selected images and the query. For better accuracy in searching, two level multiresolution matching may also be used. Masks are used for partial-sketch queries. This technique performs much better in capturing coherence of image, object granularity, local color/texture, and bias avoidance than traditional color layout algorithms. When tested on a database of more than 10,000 general-purpose images, WBIIS is much faster and more accurate than traditional algorithms.",
author = "James Wang and Gio Wiederhold and Oscar Firschein and Wei, {Sha Xin}",
year = "1997",
month = "1",
day = "1",
language = "English (US)",
pages = "13--24",
note = "Proceedings of the 1997 IEEE International Forum on Research and Technology Advances in Digital Libraries, ADL'97 ; Conference date: 07-05-1997 Through 09-05-1997",

}

Wang, J, Wiederhold, G, Firschein, O & Wei, SX 1997, 'Wavelet-based image indexing techniques with partial sketch retrieval capability' Paper presented at Proceedings of the 1997 IEEE International Forum on Research and Technology Advances in Digital Libraries, ADL'97, Washington, DC, USA, 5/7/97 - 5/9/97, pp. 13-24.

Wavelet-based image indexing techniques with partial sketch retrieval capability. / Wang, James; Wiederhold, Gio; Firschein, Oscar; Wei, Sha Xin.

1997. 13-24 Paper presented at Proceedings of the 1997 IEEE International Forum on Research and Technology Advances in Digital Libraries, ADL'97, Washington, DC, USA, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Wavelet-based image indexing techniques with partial sketch retrieval capability

AU - Wang, James

AU - Wiederhold, Gio

AU - Firschein, Oscar

AU - Wei, Sha Xin

PY - 1997/1/1

Y1 - 1997/1/1

N2 - This paper describes WBIIS (Wavelet-Based Image Indexing and Searching), a new image indexing and retrieval algorithm with partial sketch image searching capability for large image databases. The algorithm characterizes the color variations over the spatial extent of the image in a manner that provides semantically-meaningful image comparisons. The indexing algorithm applies a Daubechies' wavelet transform for each of the three opponent color components. The wavelet coefficients in the lowest few frequency bands, and their variances, are stored as feature vectors. To speed up retrieval, a two-step procedure is used that first does a crude selection based on the variances, and then refines the search by performing a feature vector match between the selected images and the query. For better accuracy in searching, two level multiresolution matching may also be used. Masks are used for partial-sketch queries. This technique performs much better in capturing coherence of image, object granularity, local color/texture, and bias avoidance than traditional color layout algorithms. When tested on a database of more than 10,000 general-purpose images, WBIIS is much faster and more accurate than traditional algorithms.

AB - This paper describes WBIIS (Wavelet-Based Image Indexing and Searching), a new image indexing and retrieval algorithm with partial sketch image searching capability for large image databases. The algorithm characterizes the color variations over the spatial extent of the image in a manner that provides semantically-meaningful image comparisons. The indexing algorithm applies a Daubechies' wavelet transform for each of the three opponent color components. The wavelet coefficients in the lowest few frequency bands, and their variances, are stored as feature vectors. To speed up retrieval, a two-step procedure is used that first does a crude selection based on the variances, and then refines the search by performing a feature vector match between the selected images and the query. For better accuracy in searching, two level multiresolution matching may also be used. Masks are used for partial-sketch queries. This technique performs much better in capturing coherence of image, object granularity, local color/texture, and bias avoidance than traditional color layout algorithms. When tested on a database of more than 10,000 general-purpose images, WBIIS is much faster and more accurate than traditional algorithms.

UR - http://www.scopus.com/inward/record.url?scp=0030683871&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030683871&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0030683871

SP - 13

EP - 24

ER -

Wang J, Wiederhold G, Firschein O, Wei SX. Wavelet-based image indexing techniques with partial sketch retrieval capability. 1997. Paper presented at Proceedings of the 1997 IEEE International Forum on Research and Technology Advances in Digital Libraries, ADL'97, Washington, DC, USA, .