IDENTIFICATION OF MARKS ON TIRES USING ARTIFICIAL VISION FOR QUALITY CONTROL
Tire inspection is presently done by workers who have as their main problems, besides identifying the defects, the time available for defect identification and the inherent costs. Companies can become more sustainable by adopting automated methods to perform such type of processes, such as artificia...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Center for Quality, Faculty of Engineering, University of Kragujevac, Serbia
2015-03-01
|
Series: | International Journal for Quality Research |
Subjects: | |
Online Access: | http://www.ijqr.net/journal/v9-n1/2.pdf |
id |
doaj-836d966639dd481c9eb3aca7e725171c |
---|---|
record_format |
Article |
spelling |
doaj-836d966639dd481c9eb3aca7e725171c2021-03-02T03:53:20ZengCenter for Quality, Faculty of Engineering, University of Kragujevac, SerbiaInternational Journal for Quality Research1800-64501800-74732015-03-01912736IDENTIFICATION OF MARKS ON TIRES USING ARTIFICIAL VISION FOR QUALITY CONTROLAndré P. Dias0Manuel F. Silva1Nuno Lima2Ricardo Guedes3INESC TEC - INESC Technology and Science (formerly INESC Porto) ISEP/IPP - School of Engineering, Polytechnic Institute of Porto Department of Electrical Engineering Porto PortugalINESC TEC - INESC Technology and Science (formerly INESC Porto) ISEP/IPP - School of Engineering, Polytechnic Institute of Porto Rua Dr. António Bernardino de Almeida Porto PortugalISEP/IPP - School of Engineering, Polytechnic Institute of Porto Rua Dr. António Bernardino de Almeida Porto PortugalISEP/IPP - School of Engineering, Polytechnic Institute of Porto Rua Dr. António Bernardino de Almeida Porto PortugalTire inspection is presently done by workers who have as their main problems, besides identifying the defects, the time available for defect identification and the inherent costs. Companies can become more sustainable by adopting automated methods to perform such type of processes, such as artificial vision, with advantages both in the processing time and in the incurred costs. This paper addresses the development of an artificial vision system that aims to be an asset in the field of tyre inspection, having as main characteristics its execution speed and its reliability. The conjugation of these criteria is a prerequisite for this system to be able to be integrated in inspection machines. The paper focusses on the study of three image processing methods to be used in the identification of marks (red dots) on tires. In this work was used the free Open Computer Vision artificial vision library to process the images acquired by a Basler matrix camera. Two different techniques, namely Background Subtraction and Hough Transform, were tested to implement the solution. After developing the artificial vision inspection application, tests were made to measure the performance of both methods and the results were promising: processing time was low and, simultaneous, the achieved accuracy is high.http://www.ijqr.net/journal/v9-n1/2.pdfTire inspectionComputer VisionImage ProcessHough TransformBackground Subtraction |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
André P. Dias Manuel F. Silva Nuno Lima Ricardo Guedes |
spellingShingle |
André P. Dias Manuel F. Silva Nuno Lima Ricardo Guedes IDENTIFICATION OF MARKS ON TIRES USING ARTIFICIAL VISION FOR QUALITY CONTROL International Journal for Quality Research Tire inspection Computer Vision Image Process Hough Transform Background Subtraction |
author_facet |
André P. Dias Manuel F. Silva Nuno Lima Ricardo Guedes |
author_sort |
André P. Dias |
title |
IDENTIFICATION OF MARKS ON TIRES USING ARTIFICIAL VISION FOR QUALITY CONTROL |
title_short |
IDENTIFICATION OF MARKS ON TIRES USING ARTIFICIAL VISION FOR QUALITY CONTROL |
title_full |
IDENTIFICATION OF MARKS ON TIRES USING ARTIFICIAL VISION FOR QUALITY CONTROL |
title_fullStr |
IDENTIFICATION OF MARKS ON TIRES USING ARTIFICIAL VISION FOR QUALITY CONTROL |
title_full_unstemmed |
IDENTIFICATION OF MARKS ON TIRES USING ARTIFICIAL VISION FOR QUALITY CONTROL |
title_sort |
identification of marks on tires using artificial vision for quality control |
publisher |
Center for Quality, Faculty of Engineering, University of Kragujevac, Serbia |
series |
International Journal for Quality Research |
issn |
1800-6450 1800-7473 |
publishDate |
2015-03-01 |
description |
Tire inspection is presently done by workers who have as their main problems, besides identifying the defects, the time available for defect identification and the inherent costs. Companies can become more sustainable by adopting automated methods to perform such type of processes, such as artificial vision, with advantages both in the processing time and in the incurred costs. This paper addresses the development of an artificial vision system that aims to be an asset in the field of tyre inspection, having as main characteristics its execution speed and its reliability. The conjugation of these criteria is a prerequisite for this system to be able to be integrated in inspection machines. The paper focusses on the study of three image processing methods to be used in the identification of marks (red dots) on tires. In this work was used the free Open Computer Vision artificial vision library to process the images acquired by a Basler matrix camera. Two different techniques, namely Background Subtraction and Hough Transform, were tested to implement the solution. After developing the artificial vision inspection application, tests were made to measure the performance of both methods and the results were promising: processing time was low and, simultaneous, the achieved accuracy is high. |
topic |
Tire inspection Computer Vision Image Process Hough Transform Background Subtraction |
url |
http://www.ijqr.net/journal/v9-n1/2.pdf |
work_keys_str_mv |
AT andrepdias identificationofmarksontiresusingartificialvisionforqualitycontrol AT manuelfsilva identificationofmarksontiresusingartificialvisionforqualitycontrol AT nunolima identificationofmarksontiresusingartificialvisionforqualitycontrol AT ricardoguedes identificationofmarksontiresusingartificialvisionforqualitycontrol |
_version_ |
1724243406966751232 |