Wavelet usage for feature extraction for crack localization

In this research work we investigate the analysis of Acoustic Emission (AE) signals using wavelet decomposition to locate a single event (crack), which usually takes place in three typical areas in a ship hull. The problem is a typical classification problem relying on the use of novel features extracted from the AE time series. As in most classification problems the extraction and selection of the most appropriate set of features plays a major role in the overall performance of the method and it is by no means a trivial task. Once a suitable set of features is extracted even ldquosimplerdquo classification models can perform adequately whereas a non-informative set of features even combined with sophisticated classifiers can lead to disappointing results. Here, we exploit the multi-resolution capabilities of wavelet decomposition, so that a set of features is extracted which it is then combined with a simple classifier. The proposed method gives superior classification rates for noisy environments compared to our previous work where conventional methods for feature extraction were deployed.