File Download
  • No File Attached
 
 
Patent History
  • Application
    EP20010109375 2001-04-18
  • Publication
    EP 1148332 2005-11-30
  • Granted
    1148332 2005-11-30
 
Supplementary

Published Patent: Method of inspecting images to detect defects
  • Basic View
  • Metadata View
  • XML View
TitleMethod of inspecting images to detect defects
 
Granted Patent1148332
 
Granted Date2005-11-30
 
Priority Date2000-04-18 US 09/197708P
 
InventorsPANG, KWOK-HUNG GRANTHAM
KUMAR, AIAY
 
Issue Date2005
 
CitationEP Published Patent application EP 1148332. European Patent Office (EPO), Escapenet, 2005 [How to Cite?]
 
AbstractThe present invention relates to a method for automated defect detection in textured materials. The present invention utilizes linear Finite Impulse Response (FIR) filters with optimized energy separation. Specifically, the invention provides a method of inspecting industrial products for defects. The method has steps of: automated design of optimized filters from samples of products, using these optimal filters to filter the acquired images of product under inspection, computing the energy of each pixel in a local region, and finally segmenting the defect by thresholding each pixel. The present invention also relates to a method of inspection of unknown (unsupervised) defects in web materials. In an unsupervised inspection, information from a finite number of optimal filters is combined using a data fusion module.; This module attempts to nullify the false alarm associated with the information arriving from different channels.
 
DC FieldValue
dc.contributor.inventorPANG, KWOK-HUNG GRANTHAM
 
dc.contributor.inventorKUMAR, AIAY
 
dc.date.accessioned2011-10-19T06:30:03Z
 
patents.date.application2001-04-18
 
dc.date.available2011-10-19T06:30:03Z
 
patents.date.granted2005-11-30
 
dc.date.issued2005
 
patents.date.priority2000-04-18 US 09/197708P
 
patents.date.publication2005-11-30
 
dc.description.abstractThe present invention relates to a method for automated defect detection in textured materials. The present invention utilizes linear Finite Impulse Response (FIR) filters with optimized energy separation. Specifically, the invention provides a method of inspecting industrial products for defects. The method has steps of: automated design of optimized filters from samples of products, using these optimal filters to filter the acquired images of product under inspection, computing the energy of each pixel in a local region, and finally segmenting the defect by thresholding each pixel. The present invention also relates to a method of inspection of unknown (unsupervised) defects in web materials. In an unsupervised inspection, information from a finite number of optimal filters is combined using a data fusion module.; This module attempts to nullify the false alarm associated with the information arriving from different channels.
 
patents.description.assigneeUNIV HONG KONG [CN]
 
patents.description.ccEP
 
patents.description.countryEuropean Patent Office
 
patents.description.kindB1
 
patents.identifier.applicationEP20010109375
 
dc.identifier.citationEP Published Patent application EP 1148332. European Patent Office (EPO), Escapenet, 2005 [How to Cite?]
 
patents.identifier.granted1148332
 
patents.identifier.hkutechidEEE-2000-00032
 
patents.identifier.publicationEP 1148332
 
dc.identifier.urihttp://hdl.handle.net/10722/142127
 
patents.relation.familyCN 1335496 (A) 2002-02-13
 
patents.relation.familyCN 100401043 (C) 2008-07-09
 
patents.relation.familyDE 60115314 (T2) 2006-08-03
 
patents.relation.familyHK 1042944 (A1) 2009-01-30
 
patents.relation.familyUS 2002054293 (A1) 2002-05-09
 
patents.relation.familyUS 6804381 (B2) 2004-10-12
 
dc.relation.isreferencedbyAT 507939 (A1) 2010-09-15
 
dc.relation.isreferencedbyAT 507939 (B1) 2012-02-15
 
dc.rightsCreative Commons: Attribution 3.0 Hong Kong License For Public Patent Documents
 
dc.titleMethod of inspecting images to detect defects
 
dc.typePatent
 
patents.typePatent_published
 
<?xml encoding="utf-8" version="1.0"?>
<item><date.accessioned>2011-10-19T06:30:03Z</date.accessioned>
<date.available>2011-10-19T06:30:03Z</date.available>
<date.issued>2005</date.issued>
<identifier.citation>EP Published Patent application EP 1148332. European Patent Office (EPO), Escapenet, 2005</identifier.citation>
<identifier.uri>http://hdl.handle.net/10722/142127</identifier.uri>
<description.abstract>The present invention relates to a method for automated defect detection in textured materials. The present invention utilizes linear Finite Impulse Response (FIR) filters with optimized energy separation. Specifically, the invention provides a method of inspecting industrial products for defects. The method has steps of: automated design of optimized filters from samples of products, using these optimal filters to filter the acquired images of product under inspection, computing the energy of each pixel in a local region, and finally segmenting the defect by thresholding each pixel. The present invention also relates to a method of inspection of unknown (unsupervised) defects in web materials. In an unsupervised inspection, information from a finite number of optimal filters is combined using a data fusion module.; This module attempts to nullify the false alarm associated with the information arriving from different channels.</description.abstract>
<relation.isreferencedby>AT 507939 (A1)  2010-09-15</relation.isreferencedby>
<relation.isreferencedby>AT 507939 (B1)  2012-02-15</relation.isreferencedby>
<rights>Creative Commons: Attribution 3.0 Hong Kong License For Public Patent Documents</rights>
<title>Method of inspecting images to detect defects</title>
<type>Patent</type>
<contributor.inventor>PANG, KWOK-HUNG GRANTHAM</contributor.inventor>
<contributor.inventor>KUMAR, AIAY</contributor.inventor>
<patents.identifier.application>EP20010109375</patents.identifier.application>
<patents.identifier.granted>1148332</patents.identifier.granted>
<patents.description.assignee>UNIV HONG KONG [CN]</patents.description.assignee>
<patents.description.country>European Patent Office</patents.description.country>
<patents.date.publication>2005-11-30</patents.date.publication>
<patents.date.granted>2005-11-30</patents.date.granted>
<patents.identifier.hkutechid>EEE-2000-00032</patents.identifier.hkutechid>
<patents.date.application>2001-04-18</patents.date.application>
<patents.date.priority>2000-04-18 US 09/197708P</patents.date.priority>
<patents.description.cc>EP</patents.description.cc>
<patents.identifier.publication>EP 1148332</patents.identifier.publication>
<patents.relation.family>CN 1335496 (A)  2002-02-13</patents.relation.family>
<patents.relation.family>CN 100401043 (C)  2008-07-09</patents.relation.family>
<patents.relation.family>DE 60115314 (T2)  2006-08-03</patents.relation.family>
<patents.relation.family>HK 1042944 (A1)  2009-01-30</patents.relation.family>
<patents.relation.family>US 2002054293 (A1)  2002-05-09</patents.relation.family>
<patents.relation.family>US 6804381 (B2)  2004-10-12</patents.relation.family>
<patents.description.kind>B1</patents.description.kind>
<patents.type>Patent_published</patents.type>
</item>