A feature selection method for automated Visual Inspection Systems

Hugo Garcia, Jesus Villalobos

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

1 Citation (Scopus)

Abstract

Automated Visual Inspection (AVI) systems are nowadays considered essential in the assembly of Surface Mounted Devices (SMD). The general goal of this research centers on developing self-training AVI systems for the inspection of SMD components. In this paper, it is proposed a new feature selection methodology based on a stepwise variable selection. The procedure uses an estimation of the marginal Misclassification Error Rate (MER) as the figure of merit to introduce new features in the quadratic classifier used by the inspection system. This marginal error rate is estimated by using the densities of the conditional stochastic representations of the underlying quadratic discriminant function. In this paper we show that the application of the proposed methodology to the inspecting of SMD components results in significant savings of computational time in the estimation of classification error over the traditional simulation and crossvalidation methods.

Original languageEnglish (US)
Title of host publicationIEEE International Conference on Industrial Informatics (INDIN)
Pages1371-1376
Number of pages6
DOIs
StatePublished - 2008
EventIEEE INDIN 2008: 6th IEEE International Conference on Industrial Informatics - Daejeon, Korea, Republic of
Duration: Jul 13 2008Jul 16 2008

Other

OtherIEEE INDIN 2008: 6th IEEE International Conference on Industrial Informatics
CountryKorea, Republic of
CityDaejeon
Period7/13/087/16/08

Fingerprint

Feature extraction
Surface mount technology
Inspection
Classifiers

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems

Cite this

Garcia, H., & Villalobos, J. (2008). A feature selection method for automated Visual Inspection Systems. In IEEE International Conference on Industrial Informatics (INDIN) (pp. 1371-1376). [4618318] https://doi.org/10.1109/INDIN.2008.4618318

A feature selection method for automated Visual Inspection Systems. / Garcia, Hugo; Villalobos, Jesus.

IEEE International Conference on Industrial Informatics (INDIN). 2008. p. 1371-1376 4618318.

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

Garcia, H & Villalobos, J 2008, A feature selection method for automated Visual Inspection Systems. in IEEE International Conference on Industrial Informatics (INDIN)., 4618318, pp. 1371-1376, IEEE INDIN 2008: 6th IEEE International Conference on Industrial Informatics, Daejeon, Korea, Republic of, 7/13/08. https://doi.org/10.1109/INDIN.2008.4618318
Garcia H, Villalobos J. A feature selection method for automated Visual Inspection Systems. In IEEE International Conference on Industrial Informatics (INDIN). 2008. p. 1371-1376. 4618318 https://doi.org/10.1109/INDIN.2008.4618318
Garcia, Hugo ; Villalobos, Jesus. / A feature selection method for automated Visual Inspection Systems. IEEE International Conference on Industrial Informatics (INDIN). 2008. pp. 1371-1376
@inproceedings{f3f840c3351e43b1992ac05a015ef902,
title = "A feature selection method for automated Visual Inspection Systems",
abstract = "Automated Visual Inspection (AVI) systems are nowadays considered essential in the assembly of Surface Mounted Devices (SMD). The general goal of this research centers on developing self-training AVI systems for the inspection of SMD components. In this paper, it is proposed a new feature selection methodology based on a stepwise variable selection. The procedure uses an estimation of the marginal Misclassification Error Rate (MER) as the figure of merit to introduce new features in the quadratic classifier used by the inspection system. This marginal error rate is estimated by using the densities of the conditional stochastic representations of the underlying quadratic discriminant function. In this paper we show that the application of the proposed methodology to the inspecting of SMD components results in significant savings of computational time in the estimation of classification error over the traditional simulation and crossvalidation methods.",
author = "Hugo Garcia and Jesus Villalobos",
year = "2008",
doi = "10.1109/INDIN.2008.4618318",
language = "English (US)",
isbn = "9781424421718",
pages = "1371--1376",
booktitle = "IEEE International Conference on Industrial Informatics (INDIN)",

}

TY - GEN

T1 - A feature selection method for automated Visual Inspection Systems

AU - Garcia, Hugo

AU - Villalobos, Jesus

PY - 2008

Y1 - 2008

N2 - Automated Visual Inspection (AVI) systems are nowadays considered essential in the assembly of Surface Mounted Devices (SMD). The general goal of this research centers on developing self-training AVI systems for the inspection of SMD components. In this paper, it is proposed a new feature selection methodology based on a stepwise variable selection. The procedure uses an estimation of the marginal Misclassification Error Rate (MER) as the figure of merit to introduce new features in the quadratic classifier used by the inspection system. This marginal error rate is estimated by using the densities of the conditional stochastic representations of the underlying quadratic discriminant function. In this paper we show that the application of the proposed methodology to the inspecting of SMD components results in significant savings of computational time in the estimation of classification error over the traditional simulation and crossvalidation methods.

AB - Automated Visual Inspection (AVI) systems are nowadays considered essential in the assembly of Surface Mounted Devices (SMD). The general goal of this research centers on developing self-training AVI systems for the inspection of SMD components. In this paper, it is proposed a new feature selection methodology based on a stepwise variable selection. The procedure uses an estimation of the marginal Misclassification Error Rate (MER) as the figure of merit to introduce new features in the quadratic classifier used by the inspection system. This marginal error rate is estimated by using the densities of the conditional stochastic representations of the underlying quadratic discriminant function. In this paper we show that the application of the proposed methodology to the inspecting of SMD components results in significant savings of computational time in the estimation of classification error over the traditional simulation and crossvalidation methods.

UR - http://www.scopus.com/inward/record.url?scp=54849424759&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=54849424759&partnerID=8YFLogxK

U2 - 10.1109/INDIN.2008.4618318

DO - 10.1109/INDIN.2008.4618318

M3 - Conference contribution

AN - SCOPUS:54849424759

SN - 9781424421718

SP - 1371

EP - 1376

BT - IEEE International Conference on Industrial Informatics (INDIN)

ER -