The USC Andrew and Erna Viterbi School of Engineering USC Signal and Image Processing Institute USC Ming Hsieh Department of Electrical Engineering University of Southern California

Technical Report USC-SIPI-221

“First Break Refraction Event Picking Using Fuzzy Logic Systems”

by Jerry M. Mendel and Chung-Kuang Chu

October 1992

First break picking is a pattern recognition problem in seismic signal processing, one that requires much human effort and is difficult to automate. Our goal is to reduce the manual effort in the picking process and accurately perform the picking.

Recently, feedforward neural network first break pickers have been developed using back propagation training algorithms applied either to an encoded version of the raw data [7] or to derived seismic attributes which are extracted from the raw data [14]. In this report we summarize a study in which we applied a recently developed back-propagation fuzzy logic system (BPFLS) to first break picking. In our work we used derived seismic attributes as features, and we also took inter-trace continuity into account by using the distance to a piecewise linear guiding function as a new feature.

Experimental results indicate that the BPFLS achieves about the same picking accuracy as a feedforward neural network that is also trained using a back propagation algorithm; however, the BPFLS is trained in a much shorter time, because of the very good way in which the initial parameters of the BPFLS can be chosen, versus the random way in which the weights of the neural network are chosen.

To download the report in PDF format click here: USC-SIPI-221.pdf (2.2Mb)