A Comparative Analysis of Machine Learning Models for First-break Arrival Picking

Abstract

First-break (FB) picking is an important and necessary step in seismic data processing and there is a need to develop precise and accurate auto-picking solutions. Our investigation in this study includes eight machine learning models. We use 1195 raw traces to extract several features and train for accurate picking and monitoring the performance of each model using well-defined evaluation metrics. Careful investigation of the scores shows that a single metric alone is not sufficient to evaluate the arrival picking models in real-time. Correlation analysis of predicted probabilities and predicted classes of machine learning models confirm that the performance metrics that use predicted probabilities have higher score value than those that use predicted classes. Our study which incorporates comparisons of different machine learning models based on different performance metrics, training time, and feature importance indicates that the approach we developed in this study is helpful and provides an opportunity to determine the real-time suitability of different methodologies for automatic FB arrival picking with clear deep insight. Based on performance scores, we bench-marked the Extra Tree classifier as the most efficient model for FB arrival picking with accuracy and F1score above 95%.

Automatic first break picking with structured random forests

“…Thus, it is of urgency to develop efficient automatic methods to quickly and accurately detect the first arrivals from complex and large‐scale seismic data. In current work, we will focus on the FBP on two‐dimensional seismic data (Ayub & Kaka, 2021; Y. Cheng et al., 2019; Coppens, 1985; Duan et al., 2018).…”

“…Over the past several decades, many advanced picking algorithms based on machine learning (especially neural networks) have been proposed to automatically or semi‐automatically detect first arrivals (Ayub & Kaka, 2021). The general procedure is to attribute the picking task as a binary classification problem (i.e., two class labels are generally taken as positive class and negative class) to solve, and the main steps can be summarized as follows.…”

Convolution neural network application for first‐break picking for land seismic data

“…Mousa et al (2011) considered the picking problem as image segmentation implemented by the method of projection onto convex sets. A huge comparison of machine learning techniques applied to the first-break picking problem was carried out by Ayub and Kaka (2021). It is important to mention an approach proposed in Turhan Taner et al (1988), based on a supervised post-picking regression analysis, that incorporates reciprocity principles.…”