19 November 2003 Neural networks based AOI systems for electronic devices diagnosis
Author Affiliations +
Proceedings Volume 4829, 19th Congress of the International Commission for Optics: Optics for the Quality of Life; (2003) https://doi.org/10.1117/12.530952
Event: 19th Congress of the International Commission for Optics: Optics for the Quality of Life, 2002, Florence, Italy
Abstract
In this work an Automatic Optical Inspection (AOI) system has been developed to diagnose Printed Circuit Boards (PCB) mounted in Surface Mounting Technology (SMT). The diagnosis task is handled as a classification problem with a neural network approach. We will present results on the diagnosis of visible defects on a SMT-PCB. A CCD camera acquires a number of images of the circuit under test and a neural network associates these images to a defect class. A set of procedures makes automatic the set-up and the diagnosis phases. The developed system seems to be a good solution in an industrial application because of the low cost, very fast diagnosis and easiness to set-up and handle.
© (2003) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Mario Lera, Mario Lera, A. Montisci, A. Montisci, "Neural networks based AOI systems for electronic devices diagnosis", Proc. SPIE 4829, 19th Congress of the International Commission for Optics: Optics for the Quality of Life, (19 November 2003); doi: 10.1117/12.530952; https://doi.org/10.1117/12.530952
PROCEEDINGS
3 PAGES


SHARE
Back to Top