Fuzzy logic based de-noising of ultrasound signals from non-destructive testing

He, Y, Manful, D, Bardossy, A, Dill-Langer, G, Ringger, T and Aicher, S (2004) Fuzzy logic based de-noising of ultrasound signals from non-destructive testing. Otto-Graf-Journal, 15. pp. 103-119.

Full text not available from this repository. (Request a copy)

Abstract

The paper reports on different methods of ultrasound signal de-noising. The reduction of noise is especially important for the evaluation of ultrasound transmission measurements in highly damping materials such as wood and glued laminated timber (glulam). In order to enable a reliable identification of characteristic signal parameters (such as time-of-flight or first amplitude) the poor signal to noise ratios (SNR) of ultrasound signals have to be improved by filtering methods. Conventional methods such as multiple signals averaging are used at the expense of huge data requirement, time consuming measurement procedures and signal processing. As an alternative approach in this paper a fuzzy logic based adaptive filter is applied for de-nosing in an attempt to use a lower number of experiments, i.e. to minimize data requirements. The results are compared to those of the conventional multiple signal averaging and of a moving average filter. Preliminary results demonstrate the feasibility of the application of the fuzzy filter and clearly illustrate its advantages as well as shortcomings over the conventional approach. The presented approach is one step towards the goal of real-time nondestructive testing (NDT) inspection of glulam beams by means of ultrasound methods.

Item Type: Article
Faculty \ School: Faculty of Science > School of Environmental Sciences
UEA Research Groups: University of East Anglia Schools > Faculty of Science > Tyndall Centre for Climate Change Research
Faculty of Science > Research Centres > Tyndall Centre for Climate Change Research
Depositing User: Rosie Cullington
Date Deposited: 25 May 2011 14:51
Last Modified: 24 Oct 2022 02:39
URI: https://ueaeprints.uea.ac.uk/id/eprint/31387
DOI:

Actions (login required)

View Item View Item