Feature generation using the Laplacian operator with Neumann boundary condition

Mohamed A. Khabou, Mohamed B H Rhouma, Lotfi Hermi

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

3 Citations (Scopus)

Abstract

The eigenvalues of the Neumann Laplacian are used to generate three different sets of features for shape recognition and classification in binary images. The generated features are rotation, translation, and size invariant and are shown to be tolerant of boundary deformation. The effectiveness of these features is demonstrated by using them to classify 5 types of computer generated and hand drawn shapes. The classification was done using 4 to 20 features fed to a simple feedforward neural network. Correct classification rates ranging from 94.4% to 100% were obtained on computer generated shapes and 67.5% to 95.5% on hand drawn shapes.

Original languageEnglish (US)
Title of host publicationConference Proceedings - IEEE SOUTHEASTCON
Pages766-771
Number of pages6
DOIs
StatePublished - 2007
Event2007 IEEE SoutheastCon - Richmond, VA, United States
Duration: Mar 22 2007Mar 25 2007

Other

Other2007 IEEE SoutheastCon
CountryUnited States
CityRichmond, VA
Period3/22/073/25/07

Fingerprint

Mathematical operators
Boundary conditions
Binary images
Feedforward neural networks

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Khabou, M. A., Rhouma, M. B. H., & Hermi, L. (2007). Feature generation using the Laplacian operator with Neumann boundary condition. In Conference Proceedings - IEEE SOUTHEASTCON (pp. 766-771). [4147535] https://doi.org/10.1109/SECON.2007.343005

Feature generation using the Laplacian operator with Neumann boundary condition. / Khabou, Mohamed A.; Rhouma, Mohamed B H; Hermi, Lotfi.

Conference Proceedings - IEEE SOUTHEASTCON. 2007. p. 766-771 4147535.

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

Khabou, MA, Rhouma, MBH & Hermi, L 2007, Feature generation using the Laplacian operator with Neumann boundary condition. in Conference Proceedings - IEEE SOUTHEASTCON., 4147535, pp. 766-771, 2007 IEEE SoutheastCon, Richmond, VA, United States, 3/22/07. https://doi.org/10.1109/SECON.2007.343005
Khabou MA, Rhouma MBH, Hermi L. Feature generation using the Laplacian operator with Neumann boundary condition. In Conference Proceedings - IEEE SOUTHEASTCON. 2007. p. 766-771. 4147535 https://doi.org/10.1109/SECON.2007.343005
Khabou, Mohamed A. ; Rhouma, Mohamed B H ; Hermi, Lotfi. / Feature generation using the Laplacian operator with Neumann boundary condition. Conference Proceedings - IEEE SOUTHEASTCON. 2007. pp. 766-771
@inproceedings{26c4cd38c33041fa8b44dfb2c9a799a4,
title = "Feature generation using the Laplacian operator with Neumann boundary condition",
abstract = "The eigenvalues of the Neumann Laplacian are used to generate three different sets of features for shape recognition and classification in binary images. The generated features are rotation, translation, and size invariant and are shown to be tolerant of boundary deformation. The effectiveness of these features is demonstrated by using them to classify 5 types of computer generated and hand drawn shapes. The classification was done using 4 to 20 features fed to a simple feedforward neural network. Correct classification rates ranging from 94.4{\%} to 100{\%} were obtained on computer generated shapes and 67.5{\%} to 95.5{\%} on hand drawn shapes.",
author = "Khabou, {Mohamed A.} and Rhouma, {Mohamed B H} and Lotfi Hermi",
year = "2007",
doi = "10.1109/SECON.2007.343005",
language = "English (US)",
isbn = "1424410290",
pages = "766--771",
booktitle = "Conference Proceedings - IEEE SOUTHEASTCON",

}

TY - GEN

T1 - Feature generation using the Laplacian operator with Neumann boundary condition

AU - Khabou, Mohamed A.

AU - Rhouma, Mohamed B H

AU - Hermi, Lotfi

PY - 2007

Y1 - 2007

N2 - The eigenvalues of the Neumann Laplacian are used to generate three different sets of features for shape recognition and classification in binary images. The generated features are rotation, translation, and size invariant and are shown to be tolerant of boundary deformation. The effectiveness of these features is demonstrated by using them to classify 5 types of computer generated and hand drawn shapes. The classification was done using 4 to 20 features fed to a simple feedforward neural network. Correct classification rates ranging from 94.4% to 100% were obtained on computer generated shapes and 67.5% to 95.5% on hand drawn shapes.

AB - The eigenvalues of the Neumann Laplacian are used to generate three different sets of features for shape recognition and classification in binary images. The generated features are rotation, translation, and size invariant and are shown to be tolerant of boundary deformation. The effectiveness of these features is demonstrated by using them to classify 5 types of computer generated and hand drawn shapes. The classification was done using 4 to 20 features fed to a simple feedforward neural network. Correct classification rates ranging from 94.4% to 100% were obtained on computer generated shapes and 67.5% to 95.5% on hand drawn shapes.

UR - http://www.scopus.com/inward/record.url?scp=34547650611&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547650611&partnerID=8YFLogxK

U2 - 10.1109/SECON.2007.343005

DO - 10.1109/SECON.2007.343005

M3 - Conference contribution

SN - 1424410290

SN - 9781424410293

SP - 766

EP - 771

BT - Conference Proceedings - IEEE SOUTHEASTCON

ER -