Y. Bashir, Nordiana MOHD MUZTAZA, Amir Abbas Babasafari, Muhammad Khan, M. Mahgoub, S.Y. Moussavi Alashloo, A. H. Abdul Latiff
{"title":"Machine Learning Application on Seismic Diffraction Detection and Preservation for High Resolution Imaging","authors":"Y. Bashir, Nordiana MOHD MUZTAZA, Amir Abbas Babasafari, Muhammad Khan, M. Mahgoub, S.Y. Moussavi Alashloo, A. H. Abdul Latiff","doi":"10.2523/iptc-21926-ea","DOIUrl":null,"url":null,"abstract":"\n Seismic Imaging for the small-scale feature in complex subsurface geology such as Carbonate is not easy to capture because of propagated waves affected by heterogeneous properties of objects in the subsurface. The initial step for machine learning (ML) is to provide enough data which can make our learning algorithm updated and mature. If one has not provided the multiple shapes of diffraction data, then your prediction of ML will be not accurate or even ML not able to detect the pattern of diffraction in the data. After the learning, our machine, the detection of the target is the crucial part that compares with the target and searches the specific signature in the given data. In this paper, we feed it with data in the form of the image and feature. Which can pass through the learning algorithm to predict the target. The idea of ML is to get the difference between your prediction and the target as closely as much possible. Which leads to the better preservation of diffractions amplitude in laterally varying velocity conditions. ML destruction is used for diffraction data separation as the conventional filtering techniques mix the diffraction amplitudes when there are a single or series of diffractions.","PeriodicalId":10974,"journal":{"name":"Day 2 Tue, February 22, 2022","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2022-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Day 2 Tue, February 22, 2022","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2523/iptc-21926-ea","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Seismic Imaging for the small-scale feature in complex subsurface geology such as Carbonate is not easy to capture because of propagated waves affected by heterogeneous properties of objects in the subsurface. The initial step for machine learning (ML) is to provide enough data which can make our learning algorithm updated and mature. If one has not provided the multiple shapes of diffraction data, then your prediction of ML will be not accurate or even ML not able to detect the pattern of diffraction in the data. After the learning, our machine, the detection of the target is the crucial part that compares with the target and searches the specific signature in the given data. In this paper, we feed it with data in the form of the image and feature. Which can pass through the learning algorithm to predict the target. The idea of ML is to get the difference between your prediction and the target as closely as much possible. Which leads to the better preservation of diffractions amplitude in laterally varying velocity conditions. ML destruction is used for diffraction data separation as the conventional filtering techniques mix the diffraction amplitudes when there are a single or series of diffractions.