Home // EMERGING 2019, The Eleventh International Conference on Emerging Networks and Systems Intelligence // View article

A Method of Feature Extraction from Time-Frequency Images of Vibration Signals in Faulty Bearings for Classification Purposes

Authors:
Dorel Aiordachioaie
Theodor Popescu
Bogdan Dumitrascu

Keywords: signal; vibration; image; time-frequency transform; signal processing; feature selection; classification

Abstract:
Time-frequency image processing is considered in the context of change detection and diagnosis purposes of bearings with faults and vibration signal processing. The analysis of these images reveals some difficulties in obtaining accurate results in classification. Some images are very different of the images from the same set. New images are obtained by applying a criterion based on the contours generated by the main components of the analyzed time-frequency image. The transformed images are less complex, and could be with white and black only. Features based on statistical moments are considered, selected and used to define discriminant functions, which could improve the results of the classification. The features include the number of the contours, the average area defined by the contours, the variance of the areas and the Renyi entropies.

Pages: 34 to 39

Copyright: Copyright (c) IARIA, 2019

Publication date: September 22, 2019

Published in: conference

ISSN: 2326-9383

ISBN: 978-1-61208-740-5

Location: Porto, Portugal

Dates: from September 22, 2019 to September 26, 2019