Invariant image recognition under projective deformations

An image normalization approach

Xue Wei, Son Lam Phung, Abdesselam Bouzerdoum, Amine Bermak

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Robustness in image recognition refers to the ability to perceive an image pattern regardless of factors including camera views and locations. This paper proposes an image normalization algorithm that allows an image with arbitrary projective distortions to be recognized efficiently. The normalization algorithm calculates the required projective transformation matrix using image moments. For an input image, a set of 8 output images that are independent of projective deformations are generated. The proposed algorithm is evaluated on three benchmark data sets. The experimental results show that the proposed normalization is significantly more accurate than the existing rank minimization and affine normalization methods.

Original languageEnglish
Title of host publication2015 Visual Communications and Image Processing, VCIP 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781467373142
DOIs
Publication statusPublished - 21 Apr 2016
Externally publishedYes
EventVisual Communications and Image Processing, VCIP 2015 - Singapore, Singapore
Duration: 13 Dec 201516 Dec 2015

Other

OtherVisual Communications and Image Processing, VCIP 2015
CountrySingapore
CitySingapore
Period13/12/1516/12/15

Fingerprint

Image recognition
Cameras

Keywords

  • geometric deformations
  • image moments
  • image normalization
  • image recognition
  • projective invariants

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Signal Processing

Cite this

Wei, X., Phung, S. L., Bouzerdoum, A., & Bermak, A. (2016). Invariant image recognition under projective deformations: An image normalization approach. In 2015 Visual Communications and Image Processing, VCIP 2015 [7457793] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VCIP.2015.7457793

Invariant image recognition under projective deformations : An image normalization approach. / Wei, Xue; Phung, Son Lam; Bouzerdoum, Abdesselam; Bermak, Amine.

2015 Visual Communications and Image Processing, VCIP 2015. Institute of Electrical and Electronics Engineers Inc., 2016. 7457793.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wei, X, Phung, SL, Bouzerdoum, A & Bermak, A 2016, Invariant image recognition under projective deformations: An image normalization approach. in 2015 Visual Communications and Image Processing, VCIP 2015., 7457793, Institute of Electrical and Electronics Engineers Inc., Visual Communications and Image Processing, VCIP 2015, Singapore, Singapore, 13/12/15. https://doi.org/10.1109/VCIP.2015.7457793
Wei X, Phung SL, Bouzerdoum A, Bermak A. Invariant image recognition under projective deformations: An image normalization approach. In 2015 Visual Communications and Image Processing, VCIP 2015. Institute of Electrical and Electronics Engineers Inc. 2016. 7457793 https://doi.org/10.1109/VCIP.2015.7457793
Wei, Xue ; Phung, Son Lam ; Bouzerdoum, Abdesselam ; Bermak, Amine. / Invariant image recognition under projective deformations : An image normalization approach. 2015 Visual Communications and Image Processing, VCIP 2015. Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{9fd7bcbf13ef45a3aa90e95a6fa53c5c,
title = "Invariant image recognition under projective deformations: An image normalization approach",
abstract = "Robustness in image recognition refers to the ability to perceive an image pattern regardless of factors including camera views and locations. This paper proposes an image normalization algorithm that allows an image with arbitrary projective distortions to be recognized efficiently. The normalization algorithm calculates the required projective transformation matrix using image moments. For an input image, a set of 8 output images that are independent of projective deformations are generated. The proposed algorithm is evaluated on three benchmark data sets. The experimental results show that the proposed normalization is significantly more accurate than the existing rank minimization and affine normalization methods.",
keywords = "geometric deformations, image moments, image normalization, image recognition, projective invariants",
author = "Xue Wei and Phung, {Son Lam} and Abdesselam Bouzerdoum and Amine Bermak",
year = "2016",
month = "4",
day = "21",
doi = "10.1109/VCIP.2015.7457793",
language = "English",
booktitle = "2015 Visual Communications and Image Processing, VCIP 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Invariant image recognition under projective deformations

T2 - An image normalization approach

AU - Wei, Xue

AU - Phung, Son Lam

AU - Bouzerdoum, Abdesselam

AU - Bermak, Amine

PY - 2016/4/21

Y1 - 2016/4/21

N2 - Robustness in image recognition refers to the ability to perceive an image pattern regardless of factors including camera views and locations. This paper proposes an image normalization algorithm that allows an image with arbitrary projective distortions to be recognized efficiently. The normalization algorithm calculates the required projective transformation matrix using image moments. For an input image, a set of 8 output images that are independent of projective deformations are generated. The proposed algorithm is evaluated on three benchmark data sets. The experimental results show that the proposed normalization is significantly more accurate than the existing rank minimization and affine normalization methods.

AB - Robustness in image recognition refers to the ability to perceive an image pattern regardless of factors including camera views and locations. This paper proposes an image normalization algorithm that allows an image with arbitrary projective distortions to be recognized efficiently. The normalization algorithm calculates the required projective transformation matrix using image moments. For an input image, a set of 8 output images that are independent of projective deformations are generated. The proposed algorithm is evaluated on three benchmark data sets. The experimental results show that the proposed normalization is significantly more accurate than the existing rank minimization and affine normalization methods.

KW - geometric deformations

KW - image moments

KW - image normalization

KW - image recognition

KW - projective invariants

UR - http://www.scopus.com/inward/record.url?scp=84978932737&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978932737&partnerID=8YFLogxK

U2 - 10.1109/VCIP.2015.7457793

DO - 10.1109/VCIP.2015.7457793

M3 - Conference contribution

BT - 2015 Visual Communications and Image Processing, VCIP 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -