Title :
On-stream defect detection of metal artifacts from line CCD image
Author :
Kurokawa, Kenta ; Morimoto, Masakazu ; Fujii, Kensaku
Author_Institution :
Div. of Comput. Eng., Univ. of Hyogo, Himeji
Abstract :
In this paper, we focus on a defect detection of a metal artifact, which has cylinder shape and hair-line finished surface, and it is sometimes ruined by scratch and/or liquid spill in factory line. So we have to find them from line CCD scanned images. In this paper, we employ Support Vector Machines (SVM) as classifier and test several transformations of input images to set input vector. We also discuss tradeoff between detection accuracy and computation time.
Keywords :
CCD image sensors; computer vision; factory automation; flaw detection; image classification; metal product industries; support vector machines; computer vision; cylinder shape; factory line; hair-line finished surface; line CCD image classifier; liquid spill; metal artifact; on-stream defect detection; support vector machine; Charge coupled devices; Engine cylinders; Fabrics; Feature extraction; Inspection; Pixel; Production facilities; Shape; Support vector machine classification; Support vector machines; Computer vision; Defect detection; Independent components analysis; Support vector machines;
Conference_Titel :
Automation Congress, 2008. WAC 2008. World
Conference_Location :
Hawaii, HI
Print_ISBN :
978-1-889335-38-4
Electronic_ISBN :
978-1-889335-37-7
