Schedule

Subscribe to RSS Feed

1975
Wednesday, January 1st
12:00 AM

Adaptive Learning Network Approach to Defect Characterization

Anthony N. Mucciardi, Adaptronics, Inc.

Thousand Oaks, CA

12:00 AM

The overall objective of this work was to demonstrate feasibility of adaptive nonlinear signal processing techniques applied to characterization of ultrasonic nondestructive testing (UNDT) waveforms for accurate inferences of flat -bottom hole sizes. The classified waveforms were ultrasonic pulse echoes obtained from two different sets of 7075-T6 aluminum area-amplitude test blocks and three different transducers. The eight flat- bottom hole defect sizes ranged from 1/64 to 8/64 inch in steps of l/64 inch

Applications of Electronically Scanned Acoustic Imaging Techniques to NDE

Gordon S. Kino, Stanford University

Thousand Oaks, CA

12:00 AM

Bob already talked about C-scan systems and I'd like to refresh your memory a little here. In particular, I want to talk about a C-scan system which has been used by Caustin on the B-1 project, in which we had considerable interest because we have been taking some of his samples and measuring them, he having measured them earlier.

Interactive Pattern Analysis and Recognition

George Forsen, Pattern Analysis and Recognition Corporation

Thousand Oaks, CA

12:00 AM

I want to stress the procedural aspects of solving pattern recognition problems. What are the tools, what are the thought processes, and what are the algorithms that you may need to do an efficient job of creating your pattern recognition design? I hope after this presentation you'll appreciate why we have spent many years in developing tools for creating pattern recognition and their designs. These are mainly computer-based, display oriented, large systems that have a comprehensive assortment of algorithms that can be easily accessed by experimenters.

Random Noise Signal Processing

V L. Newhouse, Purdue University
N M. Bilgutay, Purdue University
E S. Furgason, Purdue University

Thousand Oaks, CA

12:00 AM

Pulse echo flaw detection systems have found extensive use in industry for quality control of many types of metal and ceramic components. The random signal flaw detection system described in this paper provides an increase in sensitivity of several orders of magnitude compared to conventional pulse echo systems.

Following a review of the theory of system operation, we present some recently obtained results of our system on materials which are strongly sound absorbing, including ceramics, plastics and metals as well as material s which have large grains. In addition to detecting flaws in strongly absorbing materials we feel that this system might also be utilized as a way of estimating grain size, inclusion size or porosity.

Recent Advances in Ultrasonic Imaging

Robert C. Addison, American Optical Corporation

Thousand Oaks, CA

12:00 AM

Ultrasonics has been used for many years to detect flaws in materials by means of a pulse echo or A-scan technique. Ultrasonic images are formed by displaying the echoes in a two dimensional pattern so that t he spatial relationships between the interfaces and acoustic impedance discontinuities t hat give rise to the echoes are maintained. When the echo data is arranged in this way, it is possible for one's eye and brain to serve as a very sophisticated pattern recognition system to detect flaws or defects within an object. This capability is particularly important when one is examining an object that normally has a considerable amount of internal structure. Since most of my experience is with medical applications, I will take the human abdominal region as an example. An ultrasonic image will be able to delineate the more or less known outlines of these normal internal structures. Frequently, however, the purpose of such an examination is to look for some abnormality such as a tumor or cyst and using an ultrasonic image these structures can be differentiated from the normal structures by those who have been trained in this sort of procedure. This can be done because the operator has learned to recognize certain patterns and he is able to employ his eye-brain pattern recognition system when the echo data is presented in the form of an image.

Surface Acoustic Wave Filters for Real Time Processing of Ultrasonic Signals

Richard White, University of California - Berkeley

Thousand Oaks, CA

12:00 AM

We shall discuss using surface elastic wave device to do analog signal processing that is applicable to NOE problems . I'm going to talk today about one specific device which we have looked at which can be used with a conventional NOE system to improve resolution.

It's really brave to remain analog in an increasingly digital world, but I think there are some advantages that analog devices can have. The advantages I am referring to are speed of operation (real time operation), and the simplicity of the device, which translates, in part, into low cost.