Are Haar-like Rectangular Features for Biometric Recognition Reducible?

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

3 Citations (Scopus)
628 Downloads (Pure)

Abstract

Biometric recognition is still a very difficult task in real-world scenarios wherein unforeseen changes in degradations factors like noise, occlusion, blurriness and illumination can drastically affect the extracted features from the biometric signals. Very recently Haar-like rectangular features which have usually been used for object detection were introduced for biometric recognition resulting in systems that are robust against most of the mentioned degradations [9]. The problem with these features is that one can define many different such features for a given biometric signal and it is not clear whether all of these features are required for the actual recognition or not. This is exactly what we are dealing with in this paper: How can an initial set of Haar-like rectangular features, that have been used for biometric recognition, be reduced to a set of most influential features? This paper proposes total sensitivity analysis about the mean for this purpose for two different biometric traits, iris and face. Experimental results on multiple public databases show the superiority of the proposed system, using the found influential features, compared to state-of-the-art biometric recognition systems.
Original languageEnglish
Title of host publicationProgress in Pattern Recognition, Image Analysis, Computer Vision, and Applications
EditorsJosé Ruiz-Shulcloper, Gabriella Sanniti di Baja
Volume8259
Place of PublicationSpringer Berlin Heidelberg
PublisherSpringer Publishing Company
Publication date2013
Pages334-341
ISBN (Print)978-3-642-41826-6
ISBN (Electronic)978-3-642-41827-3
DOIs
Publication statusPublished - 2013
Event18th Iberoamerican Congress on Pattern Recognition - Havana, Cuba
Duration: 20 Nov 201323 Nov 2013

Conference

Conference18th Iberoamerican Congress on Pattern Recognition
CountryCuba
CityHavana
Period20/11/201323/11/2013
SeriesLecture Notes in Computer Science
ISSN0302-9743

Fingerprint

Biometrics
Degradation
Sensitivity analysis
Lighting

Keywords

  • Haar-like rectangular features
  • Biometric recognition

Cite this

Nasrollahi, K., & Moeslund, T. B. (2013). Are Haar-like Rectangular Features for Biometric Recognition Reducible? In J. Ruiz-Shulcloper, & G. Sanniti di Baja (Eds.), Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications (Vol. 8259, pp. 334-341). Springer Berlin Heidelberg: Springer Publishing Company. Lecture Notes in Computer Science https://doi.org/10.1007/978-3-642-41827-3_42
Nasrollahi, Kamal ; Moeslund, Thomas B. / Are Haar-like Rectangular Features for Biometric Recognition Reducible?. Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. editor / José Ruiz-Shulcloper ; Gabriella Sanniti di Baja. Vol. 8259 Springer Berlin Heidelberg : Springer Publishing Company, 2013. pp. 334-341 (Lecture Notes in Computer Science).
@inproceedings{850b1fc5a9064b33b9220812c1bc6a16,
title = "Are Haar-like Rectangular Features for Biometric Recognition Reducible?",
abstract = "Biometric recognition is still a very difficult task in real-world scenarios wherein unforeseen changes in degradations factors like noise, occlusion, blurriness and illumination can drastically affect the extracted features from the biometric signals. Very recently Haar-like rectangular features which have usually been used for object detection were introduced for biometric recognition resulting in systems that are robust against most of the mentioned degradations [9]. The problem with these features is that one can define many different such features for a given biometric signal and it is not clear whether all of these features are required for the actual recognition or not. This is exactly what we are dealing with in this paper: How can an initial set of Haar-like rectangular features, that have been used for biometric recognition, be reduced to a set of most influential features? This paper proposes total sensitivity analysis about the mean for this purpose for two different biometric traits, iris and face. Experimental results on multiple public databases show the superiority of the proposed system, using the found influential features, compared to state-of-the-art biometric recognition systems.",
keywords = "Haar-like rectangular features, Biometric recognition",
author = "Kamal Nasrollahi and Moeslund, {Thomas B.}",
year = "2013",
doi = "10.1007/978-3-642-41827-3_42",
language = "English",
isbn = "978-3-642-41826-6",
volume = "8259",
series = "Lecture Notes in Computer Science",
publisher = "Springer Publishing Company",
pages = "334--341",
editor = "Jos{\'e} Ruiz-Shulcloper and {Sanniti di Baja}, Gabriella",
booktitle = "Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications",
address = "United States",

}

Nasrollahi, K & Moeslund, TB 2013, Are Haar-like Rectangular Features for Biometric Recognition Reducible? in J Ruiz-Shulcloper & G Sanniti di Baja (eds), Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. vol. 8259, Springer Publishing Company, Springer Berlin Heidelberg, Lecture Notes in Computer Science, pp. 334-341, 18th Iberoamerican Congress on Pattern Recognition, Havana, Cuba, 20/11/2013. https://doi.org/10.1007/978-3-642-41827-3_42

Are Haar-like Rectangular Features for Biometric Recognition Reducible? / Nasrollahi, Kamal; Moeslund, Thomas B.

Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. ed. / José Ruiz-Shulcloper; Gabriella Sanniti di Baja. Vol. 8259 Springer Berlin Heidelberg : Springer Publishing Company, 2013. p. 334-341 (Lecture Notes in Computer Science).

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Are Haar-like Rectangular Features for Biometric Recognition Reducible?

AU - Nasrollahi, Kamal

AU - Moeslund, Thomas B.

PY - 2013

Y1 - 2013

N2 - Biometric recognition is still a very difficult task in real-world scenarios wherein unforeseen changes in degradations factors like noise, occlusion, blurriness and illumination can drastically affect the extracted features from the biometric signals. Very recently Haar-like rectangular features which have usually been used for object detection were introduced for biometric recognition resulting in systems that are robust against most of the mentioned degradations [9]. The problem with these features is that one can define many different such features for a given biometric signal and it is not clear whether all of these features are required for the actual recognition or not. This is exactly what we are dealing with in this paper: How can an initial set of Haar-like rectangular features, that have been used for biometric recognition, be reduced to a set of most influential features? This paper proposes total sensitivity analysis about the mean for this purpose for two different biometric traits, iris and face. Experimental results on multiple public databases show the superiority of the proposed system, using the found influential features, compared to state-of-the-art biometric recognition systems.

AB - Biometric recognition is still a very difficult task in real-world scenarios wherein unforeseen changes in degradations factors like noise, occlusion, blurriness and illumination can drastically affect the extracted features from the biometric signals. Very recently Haar-like rectangular features which have usually been used for object detection were introduced for biometric recognition resulting in systems that are robust against most of the mentioned degradations [9]. The problem with these features is that one can define many different such features for a given biometric signal and it is not clear whether all of these features are required for the actual recognition or not. This is exactly what we are dealing with in this paper: How can an initial set of Haar-like rectangular features, that have been used for biometric recognition, be reduced to a set of most influential features? This paper proposes total sensitivity analysis about the mean for this purpose for two different biometric traits, iris and face. Experimental results on multiple public databases show the superiority of the proposed system, using the found influential features, compared to state-of-the-art biometric recognition systems.

KW - Haar-like rectangular features

KW - Biometric recognition

U2 - 10.1007/978-3-642-41827-3_42

DO - 10.1007/978-3-642-41827-3_42

M3 - Article in proceeding

SN - 978-3-642-41826-6

VL - 8259

T3 - Lecture Notes in Computer Science

SP - 334

EP - 341

BT - Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications

A2 - Ruiz-Shulcloper, José

A2 - Sanniti di Baja, Gabriella

PB - Springer Publishing Company

CY - Springer Berlin Heidelberg

ER -

Nasrollahi K, Moeslund TB. Are Haar-like Rectangular Features for Biometric Recognition Reducible? In Ruiz-Shulcloper J, Sanniti di Baja G, editors, Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. Vol. 8259. Springer Berlin Heidelberg: Springer Publishing Company. 2013. p. 334-341. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-642-41827-3_42