Weld defect classification in ultrasonic testing basing on time-frequency discriminant features;

In this article we analyzed the shock response signals of a ship\'s typical part subject to non-contact underwater explosion with HHT method, calculated the Hilbert amplitude spectrum,Hilbert energy spectrum and Hilbert instantaneous energy,and discussed the time-frequency characteristics of the ships\' shock response subject to non-contact underwater explosion.

Explosive blasted per delay and burden effects on the time-frequency characteristic of vibration induced by blast

The relationship between the time-frequency characteristic of the cabin’s shock response and the underwater shock environment was obtaine.

Comparing with traditional FFT transform in analytical process,the HHT is found to be self-adaptive and superior in performance,and can effectively extract time-frequency characteristic of vibration signals of ship structures excited by the propeller vortex cavitations.

And time-frequency feature combines the time feature and the frequency feature as the final feature vector.

Since the sand body reservoir have influence on seismic reflected signals in time and frequency and wavelet transform can just reflect the time and frequency components of signals ,therefore,the pa per uses wavelet transform to analyze the time-frequency feature of seismic records.

The time-frequency features associated with the main frequency band of the motion,walking cadence and the correlation between the vertical and forward acceleration signals were combined to design a multiple classifier.

The main idea of which consists of three parts:the discriminating feature analysis of the images,the statistical modeling of face and non-face classes,and the Bayes classifier for face detection.

A novel algorithm using discriminant analysis of multiple features(DAMF)is presented for adaptive fingerprint image segmentation.