Abstract
Realistic synthetic data can be useful for data augmentation when training deep learning models to improve seismological detection and classification performance. In recent years, various deep learning techniques have been successfully applied in modern seismology. Due to the performance of deep learning depends on a sufficient volume of data, the data augmentation technique as a data-space solution is widely utilized. In this paper, we propose a Generative Adversarial Networks (GANs) based model that uses conditional knowledge to generate high-quality seismic waveforms. Unlike the existing method of generating samples directly from noise, the proposed method generates synthetic samples based on the statistical characteristics of real seismic waveforms in embedding space. Moreover, a content loss is added to relate high-level features extracted by a pre-trained model to the objective function to enhance the quality of the synthetic data. The classification accuracy is increased from 96.84% to 97.92% after mixing a certain amount of synthetic seismic waveforms, and results of the quality of seismic characteristics derived from the representative experiment show that the proposed model provides an effective structure for generating high-quality synthetic seismic waveforms. Thus, the proposed model is experimentally validated as a promising approach to realistic high-quality seismic waveform data augmentation.
Highlights
In the past hundred years alone, earthquakes have brought many disasters upon people.Earthquakes have become one of the most important issues for the growing world population and driven scientists and engineers to study them [1]
(1) We propose a novel seismic waveform generating model based on a conditional Generative adversarial networks (GANs)-based network
Traditional earthquake detection depends on the appearance of seismic waveforms
Summary
In the past hundred years alone, earthquakes have brought many disasters upon people. Machine learning incorporating deep learning has been receiving great attention in statistical seismology Deep learning algorithms, such as convolutional neural networks (CNNs), have been successful in image classification [5,6] and recurrent neural networks (RNNs) have achieved promising results with time series data [7,8]. Deep learning models using only a small dataset tend to overfit during their training process To solve this problem, data augmentation techniques have proved effective at enhancing dataset volume and quality. We propose a conditional GANs based model to generate synthetic seismic waveforms for data augmentation. The conditional GANs learn to generate the synthetic samples from a specific prior or set of characteristics rather than just random Gaussian noise. Representative experimental results demonstrate that our model is a promising approach to augmenting seismic data
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
