“Proceedings of the 2004 American Society for Engineering Education Annual Conference &
Exposition Copyright ©2004, American Society for Engineering”
Session 3447
A vision and robot based on-line monitoring of defects in Electronics
Manufacturing – A collaborative effort in capstone project.
Immanuel Edinbarough, Subhash Bose
The University of Texas at Brownsville/The University of Texas – Pan American
Abstract
This paper discusses the integration of an automated neural network-based vision
inspection system with robots to detect and report IC lead defects on-line. The vision
system consists of custom software that contains a neural network database for each
IC to be inspected on a PCB. The vision system uses gray scale images and a single
layer neural network with three outputs based on defect criteria. Each IC has different
inspection area, thus, the input vector varies for each ICs. The IC networks were
trained with Matlab’s Bayesian regularization module. This module was used because
it prevents over and under training the image data. Performance of each of the
networks investigated was found to be 100% based on the defect criteria. An on-line
robotic inspection monitoring system has been developed, using ProE, C++ and
OpenGL software
1,2
. Technical issues and collaborative efforts in the execution of
this capstone project are discussed in the paper. This research project was embarked
as a collaborative effort between the senior design project students of the University
of Texas at Brownsville and a graduate student of manufacturing engineering at the
University of Pan American.
I Introduction
Capstone design courses provide excellent opportunity for students to work on
the open ended problems that have direct bearing on the real life industry situations.
There have been several models reported in literature including the one that deal with
students work in teams on industry sponsored projects and deliver a tested prototype
at the end of the course to get credit. In this capstone project model, student teams are
formed between a graduate student team and under graduate team. The student teams
are from different colleges. There are advisors for each of the student teams from
either of the institutions. The problem that was identified for the capstone project is a
Page 9.125.1