Digital Processing and Automatic Picking of Seismic Signals

Abstract

In this research, we have implemented a numerical algorithm to identify the seismic wave parameters: P phase, S phase, amplitude, period and seismic signal duration. We have used the Matlab programming language and the digital signal processing toolbox to conduct the automatic picking of seismic signal parameters. As a mathematical tool, we have used the discrete Fourier transform to study the frequency components of the signal and the Hilbert transform to obtain the envelope signal. The Butterworth filter was used to bandpass the seismic signal, in case of a noisy signal. We have implemented a graphical user interface in Matlab to pick the seismic signals (with options of automatic picking and manual picking). In this way, the routine of picking a great set of seismic signals is easier. The results of the automatic algorithm agree very well with that of a human operator for a Signal/Noise ratio greater than 50, according to the sensitivity test, with an accuracy of around [Formula: see text]0.05[Formula: see text]s.