A new bag of words model based on fuzzy membership for image description

Yanshan Li; Weixin Xie; Zhijian Gao; Qinghua Huang; Yujie Cao

doi:10.1109/ICOSP.2014.7015149

A new bag of words model based on fuzzy membership for image description

Yanshan Li, Weixin Xie, Zhijian Gao, Qinghua Huang, Yujie Cao

Research output: Contribution to conference › Paper › peer-review

2 Scopus citations

Abstract

Bag of Words (BoW) as an efficient approach to describing the images has been attracting more and more attention. However, in traditional BoW, the maps between words in codebook and features extracted from images are ambiguous. We propose a new type of BoW based on Gaussian membership function (Gaussian-BoW) to describe images. In Gaussian-BoW, the codebook is obtained by using k-means like the traditional BoW. Then, words are assigned to the feature with Gaussian membership values. At last, histogram is generated by adding up the fuzzy membership values of each word to describe the images. The experimental results show that the proposed Gaussian-BoW outperforms traditional BoW for image description.

Original language	English
Pages	972-976
Number of pages	5
DOIs	https://doi.org/10.1109/ICOSP.2014.7015149
State	Published - 2014
Externally published	Yes
Event	2014 12th IEEE International Conference on Signal Processing, ICSP 2014 - Hangzhou, China Duration: 19 Oct 2014 → 23 Oct 2014

Conference

Conference	2014 12th IEEE International Conference on Signal Processing, ICSP 2014
Country/Territory	China
City	Hangzhou
Period	19/10/14 → 23/10/14

Keywords

Bag of words
Gaussian membership function
Image description

Access to Document

10.1109/ICOSP.2014.7015149

Cite this

@conference{bc859b64b24f49319db085803cbfb391,

title = "A new bag of words model based on fuzzy membership for image description",

abstract = "Bag of Words (BoW) as an efficient approach to describing the images has been attracting more and more attention. However, in traditional BoW, the maps between words in codebook and features extracted from images are ambiguous. We propose a new type of BoW based on Gaussian membership function (Gaussian-BoW) to describe images. In Gaussian-BoW, the codebook is obtained by using k-means like the traditional BoW. Then, words are assigned to the feature with Gaussian membership values. At last, histogram is generated by adding up the fuzzy membership values of each word to describe the images. The experimental results show that the proposed Gaussian-BoW outperforms traditional BoW for image description.",

keywords = "Bag of words, Gaussian membership function, Image description",

author = "Yanshan Li and Weixin Xie and Zhijian Gao and Qinghua Huang and Yujie Cao",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 2014 12th IEEE International Conference on Signal Processing, ICSP 2014 ; Conference date: 19-10-2014 Through 23-10-2014",

year = "2014",

doi = "10.1109/ICOSP.2014.7015149",

language = "英语",

pages = "972--976",

}

TY - CONF

T1 - A new bag of words model based on fuzzy membership for image description

AU - Li, Yanshan

AU - Xie, Weixin

AU - Gao, Zhijian

AU - Huang, Qinghua

AU - Cao, Yujie

PY - 2014

Y1 - 2014

N2 - Bag of Words (BoW) as an efficient approach to describing the images has been attracting more and more attention. However, in traditional BoW, the maps between words in codebook and features extracted from images are ambiguous. We propose a new type of BoW based on Gaussian membership function (Gaussian-BoW) to describe images. In Gaussian-BoW, the codebook is obtained by using k-means like the traditional BoW. Then, words are assigned to the feature with Gaussian membership values. At last, histogram is generated by adding up the fuzzy membership values of each word to describe the images. The experimental results show that the proposed Gaussian-BoW outperforms traditional BoW for image description.

AB - Bag of Words (BoW) as an efficient approach to describing the images has been attracting more and more attention. However, in traditional BoW, the maps between words in codebook and features extracted from images are ambiguous. We propose a new type of BoW based on Gaussian membership function (Gaussian-BoW) to describe images. In Gaussian-BoW, the codebook is obtained by using k-means like the traditional BoW. Then, words are assigned to the feature with Gaussian membership values. At last, histogram is generated by adding up the fuzzy membership values of each word to describe the images. The experimental results show that the proposed Gaussian-BoW outperforms traditional BoW for image description.

KW - Bag of words

KW - Gaussian membership function

KW - Image description

UR - http://www.scopus.com/inward/record.url?scp=84988269613&partnerID=8YFLogxK

U2 - 10.1109/ICOSP.2014.7015149

DO - 10.1109/ICOSP.2014.7015149

M3 - 论文

AN - SCOPUS:84988269613

SP - 972

EP - 976

T2 - 2014 12th IEEE International Conference on Signal Processing, ICSP 2014

Y2 - 19 October 2014 through 23 October 2014

ER -

A new bag of words model based on fuzzy membership for image description

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this