{"title":"用于减弱震荡数据振铃效应的深度 CNN 模型","authors":"Zhuang Jia, Wenkai Lu","doi":"arxiv-2408.01831","DOIUrl":null,"url":null,"abstract":"In the field of exploration geophysics, seismic vibrator is one of the widely\nused seismic sources to acquire seismic data, which is usually named vibroseis.\n\"Ringing effect\" is a common problem in vibroseis data processing due to the\nlimited frequency bandwidth of the vibrator, which degrades the performance of\nfirst-break picking. In this paper, we proposed a novel deringing model for\nvibroseis data using deep convolutional neural network (CNN). In this model we\nuse end-to-end training strategy to obtain the deringed data directly, and skip\nconnections to improve model training process and preserve the details of\nvibroseis data. For real vibroseis deringing task we synthesize training data\nand corresponding labels from real vibroseis data and utilize them to train the\ndeep CNN model. Experiments are conducted both on synthetic data and real\nvibroseis data. The experiment results show that deep CNN model can attenuate\nthe ringing effect effectively and expand the bandwidth of vibroseis data. The\nSTA/LTA ratio method for first-break picking also shows improvement on deringed\nvibroseis data using deep CNN model.","PeriodicalId":501270,"journal":{"name":"arXiv - PHYS - Geophysics","volume":"186 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Deep CNN Model for Ringing Effect Attenuation of Vibroseis Data\",\"authors\":\"Zhuang Jia, Wenkai Lu\",\"doi\":\"arxiv-2408.01831\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In the field of exploration geophysics, seismic vibrator is one of the widely\\nused seismic sources to acquire seismic data, which is usually named vibroseis.\\n\\\"Ringing effect\\\" is a common problem in vibroseis data processing due to the\\nlimited frequency bandwidth of the vibrator, which degrades the performance of\\nfirst-break picking. In this paper, we proposed a novel deringing model for\\nvibroseis data using deep convolutional neural network (CNN). In this model we\\nuse end-to-end training strategy to obtain the deringed data directly, and skip\\nconnections to improve model training process and preserve the details of\\nvibroseis data. For real vibroseis deringing task we synthesize training data\\nand corresponding labels from real vibroseis data and utilize them to train the\\ndeep CNN model. Experiments are conducted both on synthetic data and real\\nvibroseis data. The experiment results show that deep CNN model can attenuate\\nthe ringing effect effectively and expand the bandwidth of vibroseis data. The\\nSTA/LTA ratio method for first-break picking also shows improvement on deringed\\nvibroseis data using deep CNN model.\",\"PeriodicalId\":501270,\"journal\":{\"name\":\"arXiv - PHYS - Geophysics\",\"volume\":\"186 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - PHYS - Geophysics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2408.01831\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Geophysics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2408.01831","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A Deep CNN Model for Ringing Effect Attenuation of Vibroseis Data
In the field of exploration geophysics, seismic vibrator is one of the widely
used seismic sources to acquire seismic data, which is usually named vibroseis.
"Ringing effect" is a common problem in vibroseis data processing due to the
limited frequency bandwidth of the vibrator, which degrades the performance of
first-break picking. In this paper, we proposed a novel deringing model for
vibroseis data using deep convolutional neural network (CNN). In this model we
use end-to-end training strategy to obtain the deringed data directly, and skip
connections to improve model training process and preserve the details of
vibroseis data. For real vibroseis deringing task we synthesize training data
and corresponding labels from real vibroseis data and utilize them to train the
deep CNN model. Experiments are conducted both on synthetic data and real
vibroseis data. The experiment results show that deep CNN model can attenuate
the ringing effect effectively and expand the bandwidth of vibroseis data. The
STA/LTA ratio method for first-break picking also shows improvement on deringed
vibroseis data using deep CNN model.