An efficient color descriptor based on global and local color features for image retrieval

Atoany N. Fierro-Radilla, Mariko Nakano-Miyatake, Hector Perez-Meana, Manuel Cedillo-Hernandez, Francisco Garcia-Ugalde

Research output: Contribution to conferencePaper

10 Citations (Scopus)

Abstract

In the context of content-based image retrieval (CBIR) problem, an image is represented by feature vectors called descriptors, whose efficiency is essential to obtain a good performance in the image indexing and retrieval tasks. In this paper, we propose an algorithm to obtain an efficient color-based descriptor, which is a combination of Color Correlogram (CC) and Dominant Color (DC). The color-based descriptors, such as Histogram Intersection (HI) and DC take into account the global distribution of color in an image, while CC takes into account the local color distribution. So the combination of global and local color distribution provides a good image description. By its design, the proposed descriptor is more compact compared with the CC descriptor, which allows reducing computational complexity. Using the Average Retrieval precision (ARP) with different factors the effectiveness of the proposed descriptor is evaluated and compared with the conventional color-based descriptors, such as HI, CC and DC. The image database used in this work contains 500 images with 25 categories randomly selected from the Corel Dataset. © 2013 IEEE.
Original languageAmerican English
Pages233-238
Number of pages209
DOIs
StatePublished - 1 Jan 2013
Event2013 10th International Conference on Electrical Engineering, Computing Science and Automatic Control, CCE 2013 -
Duration: 1 Jan 2013 → …

Conference

Conference2013 10th International Conference on Electrical Engineering, Computing Science and Automatic Control, CCE 2013
Period1/01/13 → …

Fingerprint

Image retrieval
Color
histogram
Computational complexity

Cite this

Fierro-Radilla, A. N., Nakano-Miyatake, M., Perez-Meana, H., Cedillo-Hernandez, M., & Garcia-Ugalde, F. (2013). An efficient color descriptor based on global and local color features for image retrieval. 233-238. Paper presented at 2013 10th International Conference on Electrical Engineering, Computing Science and Automatic Control, CCE 2013, . https://doi.org/10.1109/ICEEE.2013.6676028
Fierro-Radilla, Atoany N. ; Nakano-Miyatake, Mariko ; Perez-Meana, Hector ; Cedillo-Hernandez, Manuel ; Garcia-Ugalde, Francisco. / An efficient color descriptor based on global and local color features for image retrieval. Paper presented at 2013 10th International Conference on Electrical Engineering, Computing Science and Automatic Control, CCE 2013, .209 p.
@conference{256dba965d064c2a89bb11f622329c13,
title = "An efficient color descriptor based on global and local color features for image retrieval",
abstract = "In the context of content-based image retrieval (CBIR) problem, an image is represented by feature vectors called descriptors, whose efficiency is essential to obtain a good performance in the image indexing and retrieval tasks. In this paper, we propose an algorithm to obtain an efficient color-based descriptor, which is a combination of Color Correlogram (CC) and Dominant Color (DC). The color-based descriptors, such as Histogram Intersection (HI) and DC take into account the global distribution of color in an image, while CC takes into account the local color distribution. So the combination of global and local color distribution provides a good image description. By its design, the proposed descriptor is more compact compared with the CC descriptor, which allows reducing computational complexity. Using the Average Retrieval precision (ARP) with different factors the effectiveness of the proposed descriptor is evaluated and compared with the conventional color-based descriptors, such as HI, CC and DC. The image database used in this work contains 500 images with 25 categories randomly selected from the Corel Dataset. {\circledC} 2013 IEEE.",
author = "Fierro-Radilla, {Atoany N.} and Mariko Nakano-Miyatake and Hector Perez-Meana and Manuel Cedillo-Hernandez and Francisco Garcia-Ugalde",
year = "2013",
month = "1",
day = "1",
doi = "10.1109/ICEEE.2013.6676028",
language = "American English",
pages = "233--238",
note = "2013 10th International Conference on Electrical Engineering, Computing Science and Automatic Control, CCE 2013 ; Conference date: 01-01-2013",

}

Fierro-Radilla, AN, Nakano-Miyatake, M, Perez-Meana, H, Cedillo-Hernandez, M & Garcia-Ugalde, F 2013, 'An efficient color descriptor based on global and local color features for image retrieval', Paper presented at 2013 10th International Conference on Electrical Engineering, Computing Science and Automatic Control, CCE 2013, 1/01/13 pp. 233-238. https://doi.org/10.1109/ICEEE.2013.6676028

An efficient color descriptor based on global and local color features for image retrieval. / Fierro-Radilla, Atoany N.; Nakano-Miyatake, Mariko; Perez-Meana, Hector; Cedillo-Hernandez, Manuel; Garcia-Ugalde, Francisco.

2013. 233-238 Paper presented at 2013 10th International Conference on Electrical Engineering, Computing Science and Automatic Control, CCE 2013, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - An efficient color descriptor based on global and local color features for image retrieval

AU - Fierro-Radilla, Atoany N.

AU - Nakano-Miyatake, Mariko

AU - Perez-Meana, Hector

AU - Cedillo-Hernandez, Manuel

AU - Garcia-Ugalde, Francisco

PY - 2013/1/1

Y1 - 2013/1/1

N2 - In the context of content-based image retrieval (CBIR) problem, an image is represented by feature vectors called descriptors, whose efficiency is essential to obtain a good performance in the image indexing and retrieval tasks. In this paper, we propose an algorithm to obtain an efficient color-based descriptor, which is a combination of Color Correlogram (CC) and Dominant Color (DC). The color-based descriptors, such as Histogram Intersection (HI) and DC take into account the global distribution of color in an image, while CC takes into account the local color distribution. So the combination of global and local color distribution provides a good image description. By its design, the proposed descriptor is more compact compared with the CC descriptor, which allows reducing computational complexity. Using the Average Retrieval precision (ARP) with different factors the effectiveness of the proposed descriptor is evaluated and compared with the conventional color-based descriptors, such as HI, CC and DC. The image database used in this work contains 500 images with 25 categories randomly selected from the Corel Dataset. © 2013 IEEE.

AB - In the context of content-based image retrieval (CBIR) problem, an image is represented by feature vectors called descriptors, whose efficiency is essential to obtain a good performance in the image indexing and retrieval tasks. In this paper, we propose an algorithm to obtain an efficient color-based descriptor, which is a combination of Color Correlogram (CC) and Dominant Color (DC). The color-based descriptors, such as Histogram Intersection (HI) and DC take into account the global distribution of color in an image, while CC takes into account the local color distribution. So the combination of global and local color distribution provides a good image description. By its design, the proposed descriptor is more compact compared with the CC descriptor, which allows reducing computational complexity. Using the Average Retrieval precision (ARP) with different factors the effectiveness of the proposed descriptor is evaluated and compared with the conventional color-based descriptors, such as HI, CC and DC. The image database used in this work contains 500 images with 25 categories randomly selected from the Corel Dataset. © 2013 IEEE.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84899081451&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=84899081451&origin=inward

U2 - 10.1109/ICEEE.2013.6676028

DO - 10.1109/ICEEE.2013.6676028

M3 - Paper

SP - 233

EP - 238

ER -

Fierro-Radilla AN, Nakano-Miyatake M, Perez-Meana H, Cedillo-Hernandez M, Garcia-Ugalde F. An efficient color descriptor based on global and local color features for image retrieval. 2013. Paper presented at 2013 10th International Conference on Electrical Engineering, Computing Science and Automatic Control, CCE 2013, . https://doi.org/10.1109/ICEEE.2013.6676028