Weighted LDA Based Dimensionality Reduction for Ultrasonic Flaw Signal Classification

Peng Yang; Yang Yang Tian

doi:10.4028/www.scientific.net/AMM.433-435.1841

Paper Titles

Quality Analysis Method of Tracking Information in Aerospace Flight Mission
p.1821

Transformer Fault Diagnosis Study Based on Bayesian Case Library
p.1826

Research of Low Voltage Grid Harmonic Analysis Based on DSP
p.1832

Research on Pipeline Gas Application System
p.1836

Weighted LDA Based Dimensionality Reduction for Ultrasonic Flaw Signal Classification
p.1841

Application of Information Technology in Vocational Engineering Training
p.1845

Component-Based FPGA Circuit Design and Verification for Robotic Systems Using JavaRock and ORB Engine - A Case Study
p.1849

The Agricultural Irrigation District Information System Based on Multi-Agent and GSM
p.1853

Design, Modeling and Simulation of a Novel Omni-Directional Steering System
p.1857

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 433-435Weighted LDA Based Dimensionality Reduction for...

Weighted LDA Based Dimensionality Reduction for Ultrasonic Flaw Signal Classification

Abstract:

Ultrasonic inspection is the most successful non-destructive testing technique for detection of flaws in engineering materials. Generally, discrete wavelet transform (DWT) is widely used to extract features for ultrasonic flaw signal. Directly taking the DWT coefficients as input vectors for training classifier, however, may result in computation complexity or even poor classification performanc. We propose a weighted linear discriminant analysis (WLDA) method to address the problem. In this study, DWT is first applied for the time-frequency analysis of ultrasonic flaw signals, and the wavelet coefficients are extracted as the initial features. After that, the proposed WLDA, which can effectively estimate the within-class and the between-class scatter through calculating similarity based weighting function, is used to reduce the dimension of original features. Finally, the features in new lower dimensional space are taken for flaw classification. Experimental results show that compared with the original and state of art linear discriminant analysis methods, WLDA is helpful for improving classification performance.