Interactive Ultrasonic Nondestructive Evaluation

Abstract

It has been shown that modern DSP (Digital Signal Processing) and Pattern Recognition techniques are very effective tools for Ultrasonic NDE (Non- Destructive Evaluations) of materials. In our recent researches 1, such techniques as High Resolution Spectral Analysis, Deconvolution, K-mean Clustering, Non-linear Mapping and Multipe Fisher's Discriminant, have been applied to the classification of hidden defect geometries with high accuracy. In view of this uprising needs for DSP and Pattern Recognition in Ultrasonic NDE, and software package that effectively combines the two is in order. Interactive Ultrasonic NDE (IUNDE) TM developed by Information Research Laboratory, Inc. is such a package created especially to meet the needs of researches and practicing engineers in this field. It also includes a variety of feature extraction functions that allow the user to choose features extracted from the time series of his or her interest for classification.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 1988
Accession Number
ADA204166

Entities

People

  • Chia‐Hung Chen
  • S. K. Sin

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Composite Materials
  • Computers
  • Cross Correlation
  • Digital Signal Processing
  • Feature Extraction
  • Filtration
  • Geometry
  • Materials
  • Pattern Recognition
  • Power Spectra
  • Processing Equipment
  • Recognition
  • Signal Processing
  • Spectra
  • Test And Evaluation
  • Three Dimensional
  • Two Dimensional

Readers

  • Database Systems and Applications
  • Neural Network Machine Learning.
  • Structural Health Monitoring of Composite Structures.

Technology Areas

  • AI & ML