TY - JOUR
T1 - Vector classification of SMD images
AU - Villalobos, J. Rene
AU - Arellano, Miguel
AU - Medina, Adolfo
AU - Aguirre, Fernando
N1 - Funding Information:
The authors gratefully acknowledge support from the National Science Foundation (NSF grants DMI-0100370 and DMI-0300361) and Thomson Consumer Electronics for the realization of this research. The authors also would like to thank Vernon Dickson and Tristan Ostrowski for their input in the preparation of the final version of this paper.
PY - 2003
Y1 - 2003
N2 - The automated visual inspection of surface-mounted devices (SMD) requires the correct classification of an image as either "component present" or "component absent." The inspection system must allow the classification to be fast and reliable while also assuring that the training of the classifier is simple and not time consuming. The traditional sequential approach to classifying images, while simple to implement, presents some disadvantages, including an increase in false alarm (type I) errors and the need for a time-consuming training phase. The method presented in this paper seeks to reduce classification errors by using a vector approach for the inspection of components. An experiment using real SMD inspection data is conducted to validate the performance of the proposed vector classifier. The results of the experiment support the hypothesis that the vector-based inspection approach renders fewer errors than the sequential inspection approach.
AB - The automated visual inspection of surface-mounted devices (SMD) requires the correct classification of an image as either "component present" or "component absent." The inspection system must allow the classification to be fast and reliable while also assuring that the training of the classifier is simple and not time consuming. The traditional sequential approach to classifying images, while simple to implement, presents some disadvantages, including an increase in false alarm (type I) errors and the need for a time-consuming training phase. The method presented in this paper seeks to reduce classification errors by using a vector approach for the inspection of components. An experiment using real SMD inspection data is conducted to validate the performance of the proposed vector classifier. The results of the experiment support the hypothesis that the vector-based inspection approach renders fewer errors than the sequential inspection approach.
KW - Automated Visual Inspection (AVI)
KW - Electronics Assembly
KW - Multivariate Discrimination
KW - Vector Classifier
UR - http://www.scopus.com/inward/record.url?scp=4944241460&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=4944241460&partnerID=8YFLogxK
U2 - 10.1016/s0278-6125(03)80001-8
DO - 10.1016/s0278-6125(03)80001-8
M3 - Article
AN - SCOPUS:4944241460
SN - 0278-6125
VL - 22
SP - 265
EP - 282
JO - Journal of Manufacturing Systems
JF - Journal of Manufacturing Systems
IS - 4
ER -