Characterization of welding defects by fractal analysis of ultrasonic signals

TL;DR

This paper introduces a fractal analysis-based method for classifying welding defects from ultrasonic signals, reducing data complexity and achieving accurate classification with less computational effort.

Contribution

It applies fractal tools for preprocessing ultrasonic signals and extends linear transformation techniques for defect classification, offering a more efficient approach.

Findings

Achieves low error rates comparable to nonlinear classifiers

Reduces data points significantly through fractal analysis

Demonstrates effectiveness in welding defect classification

Abstract

In this work we apply tools developed for the study of fractal properties of time series to the problem of classifying defects in welding joints probed by ultrasonic tecniques. We employ the fractal tools in a preprocessing step, producing curves with a considerably smaller number of points than in the original signals. These curves are then used in the classification step, which is realized by applying an extension of the Karhunen-Loeve linear transformation. We show that our approach leads to small error rates, comparable with those obtained by using more time-consuming methods based on non-linear classifiers.