Image Processing for Automatic Inspection of Printed Circuit Boards

• Main Authors: Po-chin Yang, 楊博欽
• Other Authors: Ming-Yi Ju
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/68184304211047570065

碩士 === 義守大學 === 資訊工程學系碩士班 === 93 === With the progress of production technology, the circuit on a printed circuit board becomes more and more complicated. The traditionally artificial inspection methods used in the manufacturing industry of printed circuit board （PCB） are unable to reduce the production cost and satisfy the requirements of electronic industry day by day. Therefore, to find new substitutive schemes becomes the urgent task. With the maturity of digital image process technology and relevant hardware equipment, the use of automatic optical inspection （AOI） to replace the artificial inspection way to reach the best production efficiency has already been a current trend. In this study, the effort toward finding all open/short defects of a PCB has been made. After taking and binarizing the PCB image, each connected object of the binary image is separated and assigned a unique number. Besides, the end positions of each connected objects are also found and recorded. A connected object table is set up by making use of the connection relation among all ends and, then, used to find the defects of a printed circuit board.