Feature Extraction Method of Ultrasonic Signal Based on Wavelet Coefficients Clustering

ABSTRACT

Support Vector Machine can well solve the classification problem of small sample, but when the dimension of input feature vector is very large, the structure of classifier is complex, the training time is long, and the performance is decreased. To solve this problem, a feature extraction method based on wavelet coefficients clustering was proposed. All the wavelet coefficients were clustered, the energy value of wavelet coefficients in each clustering was calculated and used as the input feature vector of a classifier. The dimension of input data was greatly reduced and information of specific problem was reserved. Support Vector Machine was used to identify the defects in steel plate, experiment results showed that the proposed method has higher classification accuracy.

Keywords: