Detection and classification of material attributes-a practical application of wavelet analysis

We describe a method for classifying material properties from measurements of the Barkhausen effect, which originates from a fast magnetization of ferromagnetic materials using alternating currents. We use wavelet analysis to develop a tool box for evaluating Barkhausen measurements. The described wavelet techniques allow detection of extremely weak signals in the Barkhausen noise voltage. By using a statistical classification rule, we show that the detected structures are directly related to material properties