TY - JOUR
T1 - A three-dimensional automated visual inspection system for SMT assembly
AU - Guerra, E.
AU - Villalobos, J. R.
N1 - Funding Information:
We gratefully acknowledge the support from the US Army (contract DAAB07-95-C-D321) and NSF (Grant DMI-9502897) for the realization of this research.
Copyright:
Copyright 2004 Elsevier Science B.V., Amsterdam. All rights reserved.
PY - 2001/6
Y1 - 2001/6
N2 - The emergence of surface mount technology (SMT) in the electronics industry has created new challenges. Among these challenges is the need for more advanced and capable systems to monitor the quality of the products in the manufacturing process. Years of effort have been dedicated to the development of visual inspection units using conventional two-dimensional (2D) computer vision systems. Although these systems have proven to be a possible solution to the inspection problem, the level of quality desired for real-time inspection systems has not been attained. The difference in component dimensions, legends, colors, and reflective properties make identification and classification of surface mount devices (SMDs) a very difficult task. Since the information obtained from 2D sensors is not always sufficient to correctly infer the condition of the electronic components, a tool capable of providing an additional measurement dimension should prove very valuable. This paper describes the techniques used in developing a three-dimensional (3D) automated inspection system for SMDs. In particular, we explore four different inspection approaches. The testing and results of the system show that it is technically and economically feasible to develop a low-cost, reliable automatic visual inspection (AVI) system for printed circuit boards (PCBs) using the techniques described.
AB - The emergence of surface mount technology (SMT) in the electronics industry has created new challenges. Among these challenges is the need for more advanced and capable systems to monitor the quality of the products in the manufacturing process. Years of effort have been dedicated to the development of visual inspection units using conventional two-dimensional (2D) computer vision systems. Although these systems have proven to be a possible solution to the inspection problem, the level of quality desired for real-time inspection systems has not been attained. The difference in component dimensions, legends, colors, and reflective properties make identification and classification of surface mount devices (SMDs) a very difficult task. Since the information obtained from 2D sensors is not always sufficient to correctly infer the condition of the electronic components, a tool capable of providing an additional measurement dimension should prove very valuable. This paper describes the techniques used in developing a three-dimensional (3D) automated inspection system for SMDs. In particular, we explore four different inspection approaches. The testing and results of the system show that it is technically and economically feasible to develop a low-cost, reliable automatic visual inspection (AVI) system for printed circuit boards (PCBs) using the techniques described.
KW - Automatic visual inspection
KW - Printed circuit boards
KW - Surface mount technology
UR - http://www.scopus.com/inward/record.url?scp=0035371435&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0035371435&partnerID=8YFLogxK
U2 - 10.1016/S0360-8352(01)00016-X
DO - 10.1016/S0360-8352(01)00016-X
M3 - Article
AN - SCOPUS:0035371435
VL - 40
SP - 175
EP - 190
JO - Computers and Industrial Engineering
JF - Computers and Industrial Engineering
SN - 0360-8352
IS - 1-2
ER -