Title :
Morphological skeleton algorithm for PDP production line inspection
Author :
Ge, Renyan ; Clausi, David A.
Author_Institution :
Dept. of Syst. Design Eng., Waterloo Univ., Ont., Canada
Abstract :
Morphological skeletonization is an image processing technique that reduces complex, thick-lined images to a series of single pixel lines that accurately represent the original shapes. This procedure is especially useful to simplify automated applications requiring simple shape analysis and continuity checking by reducing the amount of redundant image data. In the semiconductor inspection field, skeletonization is a process that can be used to detect defects during plasma display panel (PDP) inspection. This paper introduces a novel morphological skeletonization algorithm developed for electrode pattern inspection of PDPs. This algorithm has been successfully integrated within a commercial machine vision system
Keywords :
computer vision; feature extraction; image matching; image representation; image thinning; inspection; mathematical morphology; plasma displays; production; semiconductor device testing; PDP inspection; PDP production line inspection; automated applications; commercial machine vision system; continuity checking; defects detection; electrode pattern inspection; feature point matching; image processing; morphological skeleton algorithm; plasma display panel; redundant image data reduction; semiconductor inspection; shape analysis; shape representation; single pixel lines; thick-lined images; Electrodes; Image analysis; Image processing; Inspection; Machine vision; Pixel; Plasma displays; Production; Shape; Skeleton;
Conference_Titel :
Electrical and Computer Engineering, 2001. Canadian Conference on
Conference_Location :
Toronto, Ont.
Print_ISBN :
0-7803-6715-4
DOI :
10.1109/CCECE.2001.933598
