|
Keywords
|
Earthquake, Waveform, Localization, Data generation, Deep learning.
|
|
Abstract
|
Due to the complexity of the earthquake occurrence process and the lack of specific patterns, accurate prediction earquake location presents several challenges. However, recent studies have shown that neural networks can play a key role in earthquake data analysis, particularly in determining the exact locations of earthquakes. In this research, aimed at improving earthquake localization, earthquake data were collected from the Iranian Seismological Center, and the waveform of each earthquake was converted into images. Subsequently, these images were labeled with spatial information (latitude, longitude, and depth), and the location of each earthquake was modeled in three dimensions using a Gaussian function. Finally, a labeled earthquake dataset was created that will be used to train neural networks.
|