Image matching using dimensionally reduced embedded earth mover's distance

Fereshteh Nayyeri, Mohammad Faidzul Nasrudin

Research output: Contribution to journalArticle

Abstract

Finding similar images to a given query image can be computed by different distance measures. One of the general distance measures is the Earth Mover's Distance (EMD). Although EMD has proven its ability to retrieve similar images in an average precision of around 95%, high execution time is its major drawback. Embedding EMD into L is a solution that solves this problem by sacrificing performance; however, it generates a heavily tailed image feature vector. We aimed to reduce the execution time of embedded EMD and increase its performance using three dimension reduction methods: sampling, sketching, and Dimension Reduction in Embedding by Adjustment in Tail (DREAT). Sampling is a method that randomly picks a small fraction of the image features. On the other hand, sketching is a distance estimation method that is based on specific summary statistics. The last method, DREAT, randomly selects an equally distributed fraction of the image features. We tested the methods on handwritten Persian digit images. Our first proposed method, sampling, reduces execution time by sacrificing the recognition performance. The sketching method outperforms sampling in the recognition, but it records higher execution time. The DREAT outperforms sampling and sketching in both the execution time and performance.

Original languageEnglish
Article number749429
JournalJournal of Applied Mathematics
Volume2013
DOIs
Publication statusPublished - 2013

Fingerprint

Image matching
Image Matching
Earth (planet)
Sketching
Sampling
Execution Time
Dimension Reduction
Tail
Adjustment
Sampling Methods
Distance Measure
Statistics
Reduction Method
Feature Vector
Digit
Three-dimension
Query

ASJC Scopus subject areas

  • Applied Mathematics

Cite this

Image matching using dimensionally reduced embedded earth mover's distance. / Nayyeri, Fereshteh; Nasrudin, Mohammad Faidzul.

In: Journal of Applied Mathematics, Vol. 2013, 749429, 2013.

Research output: Contribution to journalArticle

@article{857aa5bb627d4431a1adb03260c33be5,
title = "Image matching using dimensionally reduced embedded earth mover's distance",
abstract = "Finding similar images to a given query image can be computed by different distance measures. One of the general distance measures is the Earth Mover's Distance (EMD). Although EMD has proven its ability to retrieve similar images in an average precision of around 95{\%}, high execution time is its major drawback. Embedding EMD into L is a solution that solves this problem by sacrificing performance; however, it generates a heavily tailed image feature vector. We aimed to reduce the execution time of embedded EMD and increase its performance using three dimension reduction methods: sampling, sketching, and Dimension Reduction in Embedding by Adjustment in Tail (DREAT). Sampling is a method that randomly picks a small fraction of the image features. On the other hand, sketching is a distance estimation method that is based on specific summary statistics. The last method, DREAT, randomly selects an equally distributed fraction of the image features. We tested the methods on handwritten Persian digit images. Our first proposed method, sampling, reduces execution time by sacrificing the recognition performance. The sketching method outperforms sampling in the recognition, but it records higher execution time. The DREAT outperforms sampling and sketching in both the execution time and performance.",
author = "Fereshteh Nayyeri and Nasrudin, {Mohammad Faidzul}",
year = "2013",
doi = "10.1155/2013/749429",
language = "English",
volume = "2013",
journal = "Journal of Applied Mathematics",
issn = "1110-757X",
publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - Image matching using dimensionally reduced embedded earth mover's distance

AU - Nayyeri, Fereshteh

AU - Nasrudin, Mohammad Faidzul

PY - 2013

Y1 - 2013

N2 - Finding similar images to a given query image can be computed by different distance measures. One of the general distance measures is the Earth Mover's Distance (EMD). Although EMD has proven its ability to retrieve similar images in an average precision of around 95%, high execution time is its major drawback. Embedding EMD into L is a solution that solves this problem by sacrificing performance; however, it generates a heavily tailed image feature vector. We aimed to reduce the execution time of embedded EMD and increase its performance using three dimension reduction methods: sampling, sketching, and Dimension Reduction in Embedding by Adjustment in Tail (DREAT). Sampling is a method that randomly picks a small fraction of the image features. On the other hand, sketching is a distance estimation method that is based on specific summary statistics. The last method, DREAT, randomly selects an equally distributed fraction of the image features. We tested the methods on handwritten Persian digit images. Our first proposed method, sampling, reduces execution time by sacrificing the recognition performance. The sketching method outperforms sampling in the recognition, but it records higher execution time. The DREAT outperforms sampling and sketching in both the execution time and performance.

AB - Finding similar images to a given query image can be computed by different distance measures. One of the general distance measures is the Earth Mover's Distance (EMD). Although EMD has proven its ability to retrieve similar images in an average precision of around 95%, high execution time is its major drawback. Embedding EMD into L is a solution that solves this problem by sacrificing performance; however, it generates a heavily tailed image feature vector. We aimed to reduce the execution time of embedded EMD and increase its performance using three dimension reduction methods: sampling, sketching, and Dimension Reduction in Embedding by Adjustment in Tail (DREAT). Sampling is a method that randomly picks a small fraction of the image features. On the other hand, sketching is a distance estimation method that is based on specific summary statistics. The last method, DREAT, randomly selects an equally distributed fraction of the image features. We tested the methods on handwritten Persian digit images. Our first proposed method, sampling, reduces execution time by sacrificing the recognition performance. The sketching method outperforms sampling in the recognition, but it records higher execution time. The DREAT outperforms sampling and sketching in both the execution time and performance.

UR - http://www.scopus.com/inward/record.url?scp=84893700978&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893700978&partnerID=8YFLogxK

U2 - 10.1155/2013/749429

DO - 10.1155/2013/749429

M3 - Article

VL - 2013

JO - Journal of Applied Mathematics

JF - Journal of Applied Mathematics

SN - 1110-757X

M1 - 749429

ER -