Missing component detection on PCB using neural networks

Marzieh Mogharrebi, Anton Satria Prabuwono, Shahnorbanun Sahran, Amirhossein Aghamohammadi

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

2 Citations (Scopus)

Abstract

An automated visual inspection is needed to inspect missing components on bare Printed Circuit Board (PCB). Missing footprints on the PCB will result in lack of electronic components. Therefore, any missing footprint components on the bare PCB lead to reduced performance of electronic boards. In this study, a neural network-based automatic visual inspection system for diagnosis of missing footprints on bare PCB is presented. Five types of footprint components have been classified. The images of the board are acquired and a difference operation is applied on reference image and acquired image to determine the absence of footprints on the PCB. From each footprint component, three types of geometric features are extracted. The neural network training phase is evaluated. Finally, the experimental results are shown to represent the accuracy rate of the algorithm.

Original languageEnglish
Title of host publicationLecture Notes in Electrical Engineering
Pages387-394
Number of pages8
Volume134 LNEE
DOIs
Publication statusPublished - 2011
Event2nd International Conference of Electrical and Electronics Engineering, ICEEE 2011 - Macau
Duration: 1 Dec 20112 Dec 2011

Publication series

NameLecture Notes in Electrical Engineering
Volume134 LNEE
ISSN (Print)18761100
ISSN (Electronic)18761119

Other

Other2nd International Conference of Electrical and Electronics Engineering, ICEEE 2011
CityMacau
Period1/12/112/12/11

Fingerprint

Printed circuit boards
Neural networks
Inspection

Keywords

  • automated visual inspection system
  • feature extraction
  • neural networks
  • Printed circuit board

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this

Mogharrebi, M., Prabuwono, A. S., Sahran, S., & Aghamohammadi, A. (2011). Missing component detection on PCB using neural networks. In Lecture Notes in Electrical Engineering (Vol. 134 LNEE, pp. 387-394). (Lecture Notes in Electrical Engineering; Vol. 134 LNEE). https://doi.org/10.1007/978-3-642-25905-0_51

Missing component detection on PCB using neural networks. / Mogharrebi, Marzieh; Prabuwono, Anton Satria; Sahran, Shahnorbanun; Aghamohammadi, Amirhossein.

Lecture Notes in Electrical Engineering. Vol. 134 LNEE 2011. p. 387-394 (Lecture Notes in Electrical Engineering; Vol. 134 LNEE).

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

Mogharrebi, M, Prabuwono, AS, Sahran, S & Aghamohammadi, A 2011, Missing component detection on PCB using neural networks. in Lecture Notes in Electrical Engineering. vol. 134 LNEE, Lecture Notes in Electrical Engineering, vol. 134 LNEE, pp. 387-394, 2nd International Conference of Electrical and Electronics Engineering, ICEEE 2011, Macau, 1/12/11. https://doi.org/10.1007/978-3-642-25905-0_51
Mogharrebi M, Prabuwono AS, Sahran S, Aghamohammadi A. Missing component detection on PCB using neural networks. In Lecture Notes in Electrical Engineering. Vol. 134 LNEE. 2011. p. 387-394. (Lecture Notes in Electrical Engineering). https://doi.org/10.1007/978-3-642-25905-0_51
Mogharrebi, Marzieh ; Prabuwono, Anton Satria ; Sahran, Shahnorbanun ; Aghamohammadi, Amirhossein. / Missing component detection on PCB using neural networks. Lecture Notes in Electrical Engineering. Vol. 134 LNEE 2011. pp. 387-394 (Lecture Notes in Electrical Engineering).
@inproceedings{b9c159258cb54c30943e020f80c89794,
title = "Missing component detection on PCB using neural networks",
abstract = "An automated visual inspection is needed to inspect missing components on bare Printed Circuit Board (PCB). Missing footprints on the PCB will result in lack of electronic components. Therefore, any missing footprint components on the bare PCB lead to reduced performance of electronic boards. In this study, a neural network-based automatic visual inspection system for diagnosis of missing footprints on bare PCB is presented. Five types of footprint components have been classified. The images of the board are acquired and a difference operation is applied on reference image and acquired image to determine the absence of footprints on the PCB. From each footprint component, three types of geometric features are extracted. The neural network training phase is evaluated. Finally, the experimental results are shown to represent the accuracy rate of the algorithm.",
keywords = "automated visual inspection system, feature extraction, neural networks, Printed circuit board",
author = "Marzieh Mogharrebi and Prabuwono, {Anton Satria} and Shahnorbanun Sahran and Amirhossein Aghamohammadi",
year = "2011",
doi = "10.1007/978-3-642-25905-0_51",
language = "English",
isbn = "9783642259043",
volume = "134 LNEE",
series = "Lecture Notes in Electrical Engineering",
pages = "387--394",
booktitle = "Lecture Notes in Electrical Engineering",

}

TY - GEN

T1 - Missing component detection on PCB using neural networks

AU - Mogharrebi, Marzieh

AU - Prabuwono, Anton Satria

AU - Sahran, Shahnorbanun

AU - Aghamohammadi, Amirhossein

PY - 2011

Y1 - 2011

N2 - An automated visual inspection is needed to inspect missing components on bare Printed Circuit Board (PCB). Missing footprints on the PCB will result in lack of electronic components. Therefore, any missing footprint components on the bare PCB lead to reduced performance of electronic boards. In this study, a neural network-based automatic visual inspection system for diagnosis of missing footprints on bare PCB is presented. Five types of footprint components have been classified. The images of the board are acquired and a difference operation is applied on reference image and acquired image to determine the absence of footprints on the PCB. From each footprint component, three types of geometric features are extracted. The neural network training phase is evaluated. Finally, the experimental results are shown to represent the accuracy rate of the algorithm.

AB - An automated visual inspection is needed to inspect missing components on bare Printed Circuit Board (PCB). Missing footprints on the PCB will result in lack of electronic components. Therefore, any missing footprint components on the bare PCB lead to reduced performance of electronic boards. In this study, a neural network-based automatic visual inspection system for diagnosis of missing footprints on bare PCB is presented. Five types of footprint components have been classified. The images of the board are acquired and a difference operation is applied on reference image and acquired image to determine the absence of footprints on the PCB. From each footprint component, three types of geometric features are extracted. The neural network training phase is evaluated. Finally, the experimental results are shown to represent the accuracy rate of the algorithm.

KW - automated visual inspection system

KW - feature extraction

KW - neural networks

KW - Printed circuit board

UR - http://www.scopus.com/inward/record.url?scp=84857284287&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857284287&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-25905-0_51

DO - 10.1007/978-3-642-25905-0_51

M3 - Conference contribution

SN - 9783642259043

VL - 134 LNEE

T3 - Lecture Notes in Electrical Engineering

SP - 387

EP - 394

BT - Lecture Notes in Electrical Engineering

ER -