通过深度学习结合静止小波包变换抑制地震随机噪声

IF 0.7 4区 地球科学 Q4 GEOCHEMISTRY & GEOPHYSICS
Hua Fan, Dong-Bo Wang, Yang Zhang, Wen-Xu Wang, Tao Li
{"title":"通过深度学习结合静止小波包变换抑制地震随机噪声","authors":"Hua Fan, Dong-Bo Wang, Yang Zhang, Wen-Xu Wang, Tao Li","doi":"10.1007/s11770-024-1107-6","DOIUrl":null,"url":null,"abstract":"<p>Many traditional denoising methods, such as Gaussian filtering, tend to blur and lose details or edge information while reducing noise. The stationary wavelet packet transform is a multi-scale and multi-band analysis tool. Compared with the stationary wavelet transform, it can suppress high-frequency noise while preserving more edge details. Deep learning has significantly progressed in denoising applications. DnCNN, a residual network; FFDNet, an efficient, flexible network; U-NET, a codec network; and GAN, a generative adversative network, have better denoising effects than BM3D, the most popular conventional denoising method. Therefore, SWP_hFFDNet, a random noise attenuation network based on the stationary wavelet packet transform (SWPT) and modified FFDNet, is proposed. This network combines the advantages of SWPT, Huber norm, and FFDNet. In addition, it has three characteristics: First, SWPT is an effective feature-extraction tool that can obtain low- and high-frequency features of different scales and frequency bands. Second, because the noise level map is the input of the network, the noise removal performance of different noise levels can be improved. Third, the Huber norm can reduce the sensitivity of the network to abnormal data and enhance its robustness. The network is trained using the Adam algorithm and the BSD500 dataset, which is augmented, noised, and decomposed by SWPT. Experimental and actual data processing results show that the denoising effect of the proposed method is almost the same as those of BM3D, DnCNN, and FFDNet networks for low noise. However, for high noise, the proposed method is superior to the aforementioned networks.</p>","PeriodicalId":55500,"journal":{"name":"Applied Geophysics","volume":"14 1","pages":""},"PeriodicalIF":0.7000,"publicationDate":"2024-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Suppression of seismic random noise by deep learning combined with stationary wavelet packet transform\",\"authors\":\"Hua Fan, Dong-Bo Wang, Yang Zhang, Wen-Xu Wang, Tao Li\",\"doi\":\"10.1007/s11770-024-1107-6\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Many traditional denoising methods, such as Gaussian filtering, tend to blur and lose details or edge information while reducing noise. The stationary wavelet packet transform is a multi-scale and multi-band analysis tool. Compared with the stationary wavelet transform, it can suppress high-frequency noise while preserving more edge details. Deep learning has significantly progressed in denoising applications. DnCNN, a residual network; FFDNet, an efficient, flexible network; U-NET, a codec network; and GAN, a generative adversative network, have better denoising effects than BM3D, the most popular conventional denoising method. Therefore, SWP_hFFDNet, a random noise attenuation network based on the stationary wavelet packet transform (SWPT) and modified FFDNet, is proposed. This network combines the advantages of SWPT, Huber norm, and FFDNet. In addition, it has three characteristics: First, SWPT is an effective feature-extraction tool that can obtain low- and high-frequency features of different scales and frequency bands. Second, because the noise level map is the input of the network, the noise removal performance of different noise levels can be improved. Third, the Huber norm can reduce the sensitivity of the network to abnormal data and enhance its robustness. The network is trained using the Adam algorithm and the BSD500 dataset, which is augmented, noised, and decomposed by SWPT. Experimental and actual data processing results show that the denoising effect of the proposed method is almost the same as those of BM3D, DnCNN, and FFDNet networks for low noise. However, for high noise, the proposed method is superior to the aforementioned networks.</p>\",\"PeriodicalId\":55500,\"journal\":{\"name\":\"Applied Geophysics\",\"volume\":\"14 1\",\"pages\":\"\"},\"PeriodicalIF\":0.7000,\"publicationDate\":\"2024-06-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Applied Geophysics\",\"FirstCategoryId\":\"89\",\"ListUrlMain\":\"https://doi.org/10.1007/s11770-024-1107-6\",\"RegionNum\":4,\"RegionCategory\":\"地球科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"GEOCHEMISTRY & GEOPHYSICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Geophysics","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1007/s11770-024-1107-6","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"GEOCHEMISTRY & GEOPHYSICS","Score":null,"Total":0}
引用次数: 0

摘要

许多传统的去噪方法,如高斯滤波法,在降低噪声的同时往往会模糊和丢失细节或边缘信息。静止小波包变换是一种多尺度、多波段的分析工具。与静止小波变换相比,它可以抑制高频噪声,同时保留更多的边缘细节。深度学习在去噪应用方面取得了重大进展。与最流行的传统去噪方法 BM3D 相比,残差网络 DnCNN、高效灵活的网络 FFDNet、编解码网络 U-NET 和生成对抗网络 GAN 具有更好的去噪效果。因此,我们提出了基于静态小波包变换(SWPT)和改进的 FFDNet 的随机噪声衰减网络 SWP_hFFDNet。该网络结合了 SWPT、Huber 准则和 FFDNet 的优点。此外,它还具有三个特点:首先,SWPT 是一种有效的特征提取工具,可以获得不同尺度和频段的低频和高频特征。其次,由于网络的输入是噪声电平图,因此可以提高不同噪声电平的除噪性能。第三,Huber 准则可以降低网络对异常数据的敏感性,增强其鲁棒性。该网络使用 Adam 算法和 BSD500 数据集进行训练,并通过 SWPT 对其进行增强、噪声化和分解。实验和实际数据处理结果表明,在低噪声情况下,所提方法与 BM3D、DnCNN 和 FFDNet 网络的去噪效果几乎相同。但是,对于高噪声,提出的方法优于上述网络。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Suppression of seismic random noise by deep learning combined with stationary wavelet packet transform

Many traditional denoising methods, such as Gaussian filtering, tend to blur and lose details or edge information while reducing noise. The stationary wavelet packet transform is a multi-scale and multi-band analysis tool. Compared with the stationary wavelet transform, it can suppress high-frequency noise while preserving more edge details. Deep learning has significantly progressed in denoising applications. DnCNN, a residual network; FFDNet, an efficient, flexible network; U-NET, a codec network; and GAN, a generative adversative network, have better denoising effects than BM3D, the most popular conventional denoising method. Therefore, SWP_hFFDNet, a random noise attenuation network based on the stationary wavelet packet transform (SWPT) and modified FFDNet, is proposed. This network combines the advantages of SWPT, Huber norm, and FFDNet. In addition, it has three characteristics: First, SWPT is an effective feature-extraction tool that can obtain low- and high-frequency features of different scales and frequency bands. Second, because the noise level map is the input of the network, the noise removal performance of different noise levels can be improved. Third, the Huber norm can reduce the sensitivity of the network to abnormal data and enhance its robustness. The network is trained using the Adam algorithm and the BSD500 dataset, which is augmented, noised, and decomposed by SWPT. Experimental and actual data processing results show that the denoising effect of the proposed method is almost the same as those of BM3D, DnCNN, and FFDNet networks for low noise. However, for high noise, the proposed method is superior to the aforementioned networks.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Applied Geophysics
Applied Geophysics 地学-地球化学与地球物理
CiteScore
1.50
自引率
14.30%
发文量
912
审稿时长
2 months
期刊介绍: The journal is designed to provide an academic realm for a broad blend of academic and industry papers to promote rapid communication and exchange of ideas between Chinese and world-wide geophysicists. The publication covers the applications of geoscience, geophysics, and related disciplines in the fields of energy, resources, environment, disaster, engineering, information, military, and surveying.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信