Quantitative radioscopic profile analysis via neural networks

Thumbnail Image
Date
1993
Authors
Greenawald, Edward
Poranski, Chester
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Authors
Research Projects
Organizational Units
Journal Issue
Is Version Of
Versions
Series
Series
Review of Progress in Quantitative Nondestructive Evaluation
Center for Nondestructive Evaluation

Begun in 1973, the Review of Progress in Quantitative Nondestructive Evaluation (QNDE) is the premier international NDE meeting designed to provide an interface between research and early engineering through the presentation of current ideas and results focused on facilitating a rapid transfer to engineering development.

This site provides free, public access to papers presented at the annual QNDE conference between 1983 and 1999, and abstracts for papers presented at the conference since 2001.

Department
Abstract

Artificial neural networks have been studied over a 30 year period and are a well developed computational technology applicable to a variety of difficult problems [1]. All neural networks are simulations of neurons and synapses based upon a primitive understanding of these biological structures. The distinctive feature of these networks is that they are trainable. By various iterative schemes, a set of well characterized data can be used to create a network which will produce a correct output function of an input vector. The learning is generalized, resulting in the ability to provide correct results for input vectors not contained in the training data. The term neural network has become nearly synonymous with a particular type: the feed-forward backpropagation neural network. We will use the term network in that sense here.

Comments
Description
Keywords
Citation
DOI
Copyright
Fri Jan 01 00:00:00 UTC 1993