GPR-FWI-Py: Open-source Python software for multi-scale regularized full waveform inversion in Ground Penetrating Radar using random excitation sources

IF 4.4 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences Pub Date : 2025-02-07 DOI:10.1016/j.cageo.2025.105870

Xiangyu Wang , Hai Liu , Xu Meng , Hesong Hu

{"title":"GPR-FWI-Py: Open-source Python software for multi-scale regularized full waveform inversion in Ground Penetrating Radar using random excitation sources","authors":"Xiangyu Wang , Hai Liu , Xu Meng , Hesong Hu","doi":"10.1016/j.cageo.2025.105870","DOIUrl":null,"url":null,"abstract":"<div><div>Full Waveform Inversion (FWI) of Ground Penetrating Radar (GPR) is crucial for enhancing subsurface imaging, yet its applications often confronts computational and usability challenges. This paper introduces GPR-FWI-Py, a comprehensive 2D GPR FWI code package that addresses these challenges through a multi-scale strategy, a random excitation source strategy, and Total Variation (TV) regularization. Optimized for high-performance computing, the software is developed in pure Python, ensuring both high efficiency and accessibility. Key features include user-friendly design and readability, which empower users to easily adapt and maintain the software to meet specific project needs. Performance evaluations on layered and Over-Thrust models confirm that our strategies significantly improve FWI results. The modular architecture of GPR-FWI-Py not only simplifies the integration of the FWI algorithm into GPR imaging but also enhances adaptability by supporting the introduction of additional functionalities.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"197 ","pages":"Article 105870"},"PeriodicalIF":4.4000,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers & Geosciences","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0098300425000202","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}

引用次数: 0

Abstract

Full Waveform Inversion (FWI) of Ground Penetrating Radar (GPR) is crucial for enhancing subsurface imaging, yet its applications often confronts computational and usability challenges. This paper introduces GPR-FWI-Py, a comprehensive 2D GPR FWI code package that addresses these challenges through a multi-scale strategy, a random excitation source strategy, and Total Variation (TV) regularization. Optimized for high-performance computing, the software is developed in pure Python, ensuring both high efficiency and accessibility. Key features include user-friendly design and readability, which empower users to easily adapt and maintain the software to meet specific project needs. Performance evaluations on layered and Over-Thrust models confirm that our strategies significantly improve FWI results. The modular architecture of GPR-FWI-Py not only simplifies the integration of the FWI algorithm into GPR imaging but also enhances adaptability by supporting the introduction of additional functionalities.

查看原文本刊更多论文

GPR-FWI-Py：基于随机激励源的探地雷达多尺度正则化全波形反演的开源Python软件

探地雷达（GPR）的全波形反演（FWI）对于增强地下成像至关重要，但其应用往往面临计算和可用性方面的挑战。本文介绍了GPR-FWI- py，这是一个全面的2D GPR FWI代码包，通过多尺度策略、随机激励源策略和全变分（TV）正则化来解决这些挑战。针对高性能计算进行了优化，该软件使用纯Python开发，确保了高效率和可访问性。主要特性包括用户友好的设计和可读性，这使用户能够轻松地调整和维护软件以满足特定的项目需求。对分层和超推力模型的性能评估证实，我们的策略显著提高了FWI结果。GPR-FWI- py的模块化架构不仅简化了FWI算法与GPR成像的集成，而且通过支持引入附加功能增强了适应性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Computers & Geosciences 地学-地球科学综合

CiteScore

9.30

自引率

6.80%

发文量

164

审稿时长

3.4 months

期刊介绍： Computers & Geosciences publishes high impact, original research at the interface between Computer Sciences and Geosciences. Publications should apply modern computer science paradigms, whether computational or informatics-based, to address problems in the geosciences.