Bone texture characterization with fisher encoding of local descriptors

Yang Song; Weidong Cai; Fan Zhang; Heng Huang; Yun Zhou; David Dagan Feng

doi:10.1109/ISBI.2015.7163803

Bone texture characterization with fisher encoding of local descriptors

Yang Song, Weidong Cai, Fan Zhang, Heng Huang, Yun Zhou, David Dagan Feng

School of Medicine

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

7 Scopus citations

Abstract

Bone texture characterization is important for differentiating osteoporotic and healthy subjects. Automated classification is however very challenging due to the high degree of visual similarity between the two types of images. In this paper, we propose to describe the bone textures by extracting dense sets of local descriptors and encoding them with the improved Fisher vector (IFV). Compared to the standard bag-of-visual-words (BoW) model, Fisher encoding is more discriminative by representing the distribution of local descriptors in addition to the occurrence frequencies. Our method is evaluated on the ISBI 2014 challenge dataset of bone texture characterization, and we demonstrate excellent classification performance compared to the challenge entries and large improvement over the BoW model.

Original language	English (US)
Title of host publication	Proceedings - International Symposium on Biomedical Imaging
Publisher	IEEE Computer Society
Pages	5-8
Number of pages	4
Volume	2015-July
ISBN (Print)	9781479923748
DOIs	https://doi.org/10.1109/ISBI.2015.7163803
State	Published - Jul 21 2015
Event	12th IEEE International Symposium on Biomedical Imaging, ISBI 2015 - Brooklyn, United States Duration: Apr 16 2015 → Apr 19 2015

Other

Other	12th IEEE International Symposium on Biomedical Imaging, ISBI 2015
Country/Territory	United States
City	Brooklyn
Period	4/16/15 → 4/19/15

Keywords

Bone texture
classification
feature encoding
Fisher vector

ASJC Scopus subject areas

Biomedical Engineering
Radiology Nuclear Medicine and imaging

Access to Document

10.1109/ISBI.2015.7163803

Cite this

Song, Y, Cai, W, Zhang, F, Huang, H, Zhou, Y & Dagan Feng, D 2015, Bone texture characterization with fisher encoding of local descriptors. in Proceedings - International Symposium on Biomedical Imaging. vol. 2015-July, 7163803, IEEE Computer Society, pp. 5-8, 12th IEEE International Symposium on Biomedical Imaging, ISBI 2015, Brooklyn, United States, 4/16/15. https://doi.org/10.1109/ISBI.2015.7163803

@inproceedings{f1768799a8dd41b7bc836d176c636d86,

title = "Bone texture characterization with fisher encoding of local descriptors",

abstract = "Bone texture characterization is important for differentiating osteoporotic and healthy subjects. Automated classification is however very challenging due to the high degree of visual similarity between the two types of images. In this paper, we propose to describe the bone textures by extracting dense sets of local descriptors and encoding them with the improved Fisher vector (IFV). Compared to the standard bag-of-visual-words (BoW) model, Fisher encoding is more discriminative by representing the distribution of local descriptors in addition to the occurrence frequencies. Our method is evaluated on the ISBI 2014 challenge dataset of bone texture characterization, and we demonstrate excellent classification performance compared to the challenge entries and large improvement over the BoW model.",

keywords = "Bone texture, classification, feature encoding, Fisher vector",

author = "Yang Song and Weidong Cai and Fan Zhang and Heng Huang and Yun Zhou and {Dagan Feng}, David",

year = "2015",

month = jul,

day = "21",

doi = "10.1109/ISBI.2015.7163803",

language = "English (US)",

isbn = "9781479923748",

volume = "2015-July",

pages = "5--8",

booktitle = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE Computer Society",

note = "12th IEEE International Symposium on Biomedical Imaging, ISBI 2015 ; Conference date: 16-04-2015 Through 19-04-2015",

}

TY - GEN

T1 - Bone texture characterization with fisher encoding of local descriptors

AU - Song, Yang

AU - Cai, Weidong

AU - Zhang, Fan

AU - Huang, Heng

AU - Zhou, Yun

AU - Dagan Feng, David

PY - 2015/7/21

Y1 - 2015/7/21

N2 - Bone texture characterization is important for differentiating osteoporotic and healthy subjects. Automated classification is however very challenging due to the high degree of visual similarity between the two types of images. In this paper, we propose to describe the bone textures by extracting dense sets of local descriptors and encoding them with the improved Fisher vector (IFV). Compared to the standard bag-of-visual-words (BoW) model, Fisher encoding is more discriminative by representing the distribution of local descriptors in addition to the occurrence frequencies. Our method is evaluated on the ISBI 2014 challenge dataset of bone texture characterization, and we demonstrate excellent classification performance compared to the challenge entries and large improvement over the BoW model.

AB - Bone texture characterization is important for differentiating osteoporotic and healthy subjects. Automated classification is however very challenging due to the high degree of visual similarity between the two types of images. In this paper, we propose to describe the bone textures by extracting dense sets of local descriptors and encoding them with the improved Fisher vector (IFV). Compared to the standard bag-of-visual-words (BoW) model, Fisher encoding is more discriminative by representing the distribution of local descriptors in addition to the occurrence frequencies. Our method is evaluated on the ISBI 2014 challenge dataset of bone texture characterization, and we demonstrate excellent classification performance compared to the challenge entries and large improvement over the BoW model.

KW - Bone texture

KW - classification

KW - feature encoding

KW - Fisher vector

UR - http://www.scopus.com/inward/record.url?scp=84944325845&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944325845&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2015.7163803

DO - 10.1109/ISBI.2015.7163803

M3 - Conference contribution

AN - SCOPUS:84944325845

SN - 9781479923748

VL - 2015-July

SP - 5

EP - 8

BT - Proceedings - International Symposium on Biomedical Imaging

PB - IEEE Computer Society

T2 - 12th IEEE International Symposium on Biomedical Imaging, ISBI 2015

Y2 - 16 April 2015 through 19 April 2015

ER -

Bone texture characterization with fisher encoding of local descriptors

Abstract

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this