Scene image classification: Some novel descriptors

Sugata Banerji; Atreyee Sinha; Chengjun Liu

doi:10.1109/ICSMC.2012.6378083

Scene image classification: Some novel descriptors

Sugata Banerji, Atreyee Sinha, Chengjun Liu

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

7 Scopus citations

Abstract

This paper introduces several novel color, shape and texture-based image descriptors for scene image classification with applications to image search and retrieval. Specifically, first, a new 3-Dimensional Local Binary Pattern (3DLBP) descriptor is proposed for color image local feature extraction. Second, a new shape descriptor (HaarHOG) is introduced by combining Haar wavelet transformation and Histogram of Oriented Gradients (HOG). Third, these descriptors are fused using an optimal feature representation technique to generate a robust 3-Dimensional LBP-HaarHOG (3DLH) descriptor that can perform well on different scene image categories. Finally, the Enhanced Fisher Model (EFM) is applied for discriminatory feature extraction and the nearest neighbor classification rule is used for image classification. The proposed descriptors and fusion technique are evaluated using three grand challenge datasets: the MIT Scene dataset, the UIUC Sports Event dataset, and a part of the Caltech 256 dataset.

Original language	English (US)
Title of host publication	Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012
Pages	2294-2299
Number of pages	6
DOIs	https://doi.org/10.1109/ICSMC.2012.6378083
State	Published - 2012
Event	2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012 - Seoul, Korea, Republic of Duration: Oct 14 2012 → Oct 17 2012

Publication series

Name	Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
ISSN (Print)	1062-922X

Other

Other	2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012
Country/Territory	Korea, Republic of
City	Seoul
Period	10/14/12 → 10/17/12

All Science Journal Classification (ASJC) codes

Electrical and Electronic Engineering
Control and Systems Engineering
Human-Computer Interaction

Keywords

Enhanced Fisher Model (EFM)
The HaarHOG descriptor
image search
scene classification
the 3-Dimensional Local Binary Pattern (3D-LBP) descriptor
the 3DLH descriptor

Access to Document

10.1109/ICSMC.2012.6378083

Cite this

@inproceedings{6a574d86aa8143c39a761afc4113c02c,

title = "Scene image classification: Some novel descriptors",

abstract = "This paper introduces several novel color, shape and texture-based image descriptors for scene image classification with applications to image search and retrieval. Specifically, first, a new 3-Dimensional Local Binary Pattern (3DLBP) descriptor is proposed for color image local feature extraction. Second, a new shape descriptor (HaarHOG) is introduced by combining Haar wavelet transformation and Histogram of Oriented Gradients (HOG). Third, these descriptors are fused using an optimal feature representation technique to generate a robust 3-Dimensional LBP-HaarHOG (3DLH) descriptor that can perform well on different scene image categories. Finally, the Enhanced Fisher Model (EFM) is applied for discriminatory feature extraction and the nearest neighbor classification rule is used for image classification. The proposed descriptors and fusion technique are evaluated using three grand challenge datasets: the MIT Scene dataset, the UIUC Sports Event dataset, and a part of the Caltech 256 dataset.",

keywords = "Enhanced Fisher Model (EFM), The HaarHOG descriptor, image search, scene classification, the 3-Dimensional Local Binary Pattern (3D-LBP) descriptor, the 3DLH descriptor",

author = "Sugata Banerji and Atreyee Sinha and Chengjun Liu",

year = "2012",

doi = "10.1109/ICSMC.2012.6378083",

language = "English (US)",

isbn = "9781467317146",

series = "Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics",

pages = "2294--2299",

booktitle = "Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012",

}

Banerji, S, Sinha, A & Liu, C 2012, Scene image classification: Some novel descriptors. in Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012., 6378083, Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2294-2299, 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012, Seoul, Korea, Republic of, 10/14/12. https://doi.org/10.1109/ICSMC.2012.6378083

Scene image classification: Some novel descriptors. / Banerji, Sugata; Sinha, Atreyee; Liu, Chengjun.
Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012. 2012. p. 2294-2299 6378083 (Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Scene image classification

T2 - 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012

AU - Banerji, Sugata

AU - Sinha, Atreyee

AU - Liu, Chengjun

PY - 2012

Y1 - 2012

N2 - This paper introduces several novel color, shape and texture-based image descriptors for scene image classification with applications to image search and retrieval. Specifically, first, a new 3-Dimensional Local Binary Pattern (3DLBP) descriptor is proposed for color image local feature extraction. Second, a new shape descriptor (HaarHOG) is introduced by combining Haar wavelet transformation and Histogram of Oriented Gradients (HOG). Third, these descriptors are fused using an optimal feature representation technique to generate a robust 3-Dimensional LBP-HaarHOG (3DLH) descriptor that can perform well on different scene image categories. Finally, the Enhanced Fisher Model (EFM) is applied for discriminatory feature extraction and the nearest neighbor classification rule is used for image classification. The proposed descriptors and fusion technique are evaluated using three grand challenge datasets: the MIT Scene dataset, the UIUC Sports Event dataset, and a part of the Caltech 256 dataset.

AB - This paper introduces several novel color, shape and texture-based image descriptors for scene image classification with applications to image search and retrieval. Specifically, first, a new 3-Dimensional Local Binary Pattern (3DLBP) descriptor is proposed for color image local feature extraction. Second, a new shape descriptor (HaarHOG) is introduced by combining Haar wavelet transformation and Histogram of Oriented Gradients (HOG). Third, these descriptors are fused using an optimal feature representation technique to generate a robust 3-Dimensional LBP-HaarHOG (3DLH) descriptor that can perform well on different scene image categories. Finally, the Enhanced Fisher Model (EFM) is applied for discriminatory feature extraction and the nearest neighbor classification rule is used for image classification. The proposed descriptors and fusion technique are evaluated using three grand challenge datasets: the MIT Scene dataset, the UIUC Sports Event dataset, and a part of the Caltech 256 dataset.

KW - Enhanced Fisher Model (EFM)

KW - The HaarHOG descriptor

KW - image search

KW - scene classification

KW - the 3-Dimensional Local Binary Pattern (3D-LBP) descriptor

KW - the 3DLH descriptor

UR - http://www.scopus.com/inward/record.url?scp=84872410750&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872410750&partnerID=8YFLogxK

U2 - 10.1109/ICSMC.2012.6378083

DO - 10.1109/ICSMC.2012.6378083

M3 - Conference contribution

AN - SCOPUS:84872410750

SN - 9781467317146

T3 - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics

SP - 2294

EP - 2299

BT - Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012

Y2 - 14 October 2012 through 17 October 2012

ER -

Scene image classification: Some novel descriptors

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this