Automatic Image-Based Event Detection for Large-N Seismic Arrays Using a Convolutional Neural Network

Abstract

Passive seismic experiments have been proposed as a cost-effective and non-invasive alternative to controlled-source seismology, allowing body–wave reflections based on seismic interferometry principles to be retrieved. However, from the huge volume of the recorded ambient noise, only selected time periods (noise panels) are contributing constructively to the retrieval of reflections. We address the issue of automatic scanning of ambient noise data recorded by a large-N array in search of body–wave energy (body–wave events) utilizing a convolutional neural network (CNN). It consists of computing first both amplitude and frequency attribute values at each receiver station for all divided portions of the recorded signal (noise panels). The created 2-D attribute maps are then converted to images and used to extract spatial and temporal patterns associated with the body–wave energy present in the data to build binary CNN-based classifiers. The ensemble of two multi-headed CNN models trained separately on the frequency and amplitude attribute maps demonstrates better generalization ability than each of its participating networks. We also compare the prediction performance of our deep learning (DL) framework with a conventional machine learning (ML) algorithm called XGBoost. The DL-based solution applied to 240 h of ambient seismic noise data recorded by the Kylylahti array in Finland demonstrates high detection accuracy and the superiority over the ML-based one. The ensemble of CNN-based models managed to find almost three times more verified body–wave events in the full unlabelled dataset than it was provided at the training stage. Moreover, the high-level abstraction features extracted at the deeper convolution layers can be used to perform unsupervised clustering of the classified panels with respect to their visual characteristics.

Few-Shot Learning for Post-Earthquake Urban Damage Detection

“…In the broader context of emergency response, the significance of the damage classification output should be validated by other data sources. Seismic vulnerability reports, smart building structural monitoring [46] and earthquake magnitude measurements based on environmental noise [4] can be ancillary information to earth observation-based damage classification. The aforementioned information sources, along with ground-truth verification, should be taken into account for the response planning by the local authorities, as the traffic network capacity and the citizen mobility demands change dramatically during the recovery process [16].…”

“…Harirchian et al [3] assess the seismic vulnerability given a set of quantifiable parameters. The regional seismicity can also be monitored using ML if the automatically captured ambient noise data are subjected to Convolutional Neural Network (CNN)-based classifiers in order to detect earthquake events [4]. In the post-disaster phase, ML approaches have been employed to locate and measure post-earthquake urban damage, confirming the relevance of this field for such kinds of applications [5][6][7][8][9][10][11][12][13][14][15].…”

Optimized Ensemble Approach for Multi-model Event Detection in Big data

“…The ability to perform iterative data discovery, which results in analysis, detection, and information extraction such as popular events, is what big data most importantly permits. According to this challenge, in particularly concentrate on this work on the topic of automatic online event detection [4] on Twitter microblogs by fusing a big data analytics environment [5] with Twitter analytics to produce a novel approach that can improve event detection within the big data space. Big data demonstrated that social media data is useful for identifying data dissemination features and spotting earthquake myths.…”

A Two-Stage Seismic Damage Assessment Method for Small, Dense, and Imbalanced Buildings in Remote Sensing Images

“…This supplies a new option that utilizes computers to process data and enables computers to analyze and interpret it. In recent years, many efficient algorithms have been created to promote the application of ML and CV methods in structural health monitoring [14][15][16][17][18][19][20][21][22][23]. These methods effectively reduce or eliminate the dependence on professional measuring equipment, expensive sensors, and the subjective experience of inspectors.…”