Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data
One of the major hot topics in seismic data processing is the reconstruction of successively sampled seismic data. There are numerous traditional methods proposed for addressing this issue; however, they still have unavoidable drawbacks, such as high computational cost and sensitive tuning parameter...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10110973/ |
| _version_ | 1797680907748900864 |
|---|---|
| author | Shengjun Li Jianhu Gao Jinyong Gui Lukun Wu Naihao Liu Dongyang He Xin Guo |
| author_facet | Shengjun Li Jianhu Gao Jinyong Gui Lukun Wu Naihao Liu Dongyang He Xin Guo |
| author_sort | Shengjun Li |
| collection | DOAJ |
| description | One of the major hot topics in seismic data processing is the reconstruction of successively sampled seismic data. There are numerous traditional methods proposed for addressing this issue; however, they still have unavoidable drawbacks, such as high computational cost and sensitive tuning parameters. In this study, we suggest a deep learning model for reconstructing successively sampled seismic data, termed fully connected U-Net (FCU-Net). FCU-Net maintains the high-resolution representations by connecting the parallel different-resolution representations and repeating multi-scale fusion. Such a structure allows FCU-Net to successfully extract multi-scale information, which is beneficial for accurate seismic data reconstruction. Additionally, the extending subnetwork of FCU-Net contains a large number of feature channels and sufficient information interaction between different resolution representations via the composite cascades, which contributes to locating successively sampled traces with big gaps and then performing the seismic interpolation. To verify the effectiveness of FCU-Net, we compare it with state-of-the-art networks, i.e., U-Net and HRNet, using synthetic and field examples. The results show that FCU-Net performs best when interpolating successively sampled seismic data, proving its superiority and availability. |
| first_indexed | 2024-03-11T23:37:04Z |
| format | Article |
| id | doaj.art-96b598f941d54cfc83b16a71e51b7e88 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-11T23:37:04Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-96b598f941d54cfc83b16a71e51b7e882023-09-19T23:01:22ZengIEEEIEEE Access2169-35362023-01-0111996939970410.1109/ACCESS.2023.327151810110973Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic DataShengjun Li0https://orcid.org/0000-0001-6612-9323Jianhu Gao1https://orcid.org/0000-0001-9312-4966Jinyong Gui2Lukun Wu3Naihao Liu4https://orcid.org/0000-0002-2609-7408Dongyang He5Xin Guo6PetroChina Research Institute of Petroleum Exploration and Development, Northwest, China National Petroleum Corporation (CNPC), Lanzhou, ChinaPetroChina Research Institute of Petroleum Exploration and Development, Northwest, China National Petroleum Corporation (CNPC), Lanzhou, ChinaPetroChina Research Institute of Petroleum Exploration and Development, Northwest, China National Petroleum Corporation (CNPC), Lanzhou, ChinaSchool of Software Engineering, Xi’an Jiaotong University, Xi’an, ChinaSchool of Information and Communications Engineering, Xi’an Jiaotong University, Xi’an, ChinaPetroChina Research Institute of Petroleum Exploration and Development, Northwest, China National Petroleum Corporation (CNPC), Lanzhou, ChinaPetroChina Research Institute of Petroleum Exploration and Development, Northwest, China National Petroleum Corporation (CNPC), Lanzhou, ChinaOne of the major hot topics in seismic data processing is the reconstruction of successively sampled seismic data. There are numerous traditional methods proposed for addressing this issue; however, they still have unavoidable drawbacks, such as high computational cost and sensitive tuning parameters. In this study, we suggest a deep learning model for reconstructing successively sampled seismic data, termed fully connected U-Net (FCU-Net). FCU-Net maintains the high-resolution representations by connecting the parallel different-resolution representations and repeating multi-scale fusion. Such a structure allows FCU-Net to successfully extract multi-scale information, which is beneficial for accurate seismic data reconstruction. Additionally, the extending subnetwork of FCU-Net contains a large number of feature channels and sufficient information interaction between different resolution representations via the composite cascades, which contributes to locating successively sampled traces with big gaps and then performing the seismic interpolation. To verify the effectiveness of FCU-Net, we compare it with state-of-the-art networks, i.e., U-Net and HRNet, using synthetic and field examples. The results show that FCU-Net performs best when interpolating successively sampled seismic data, proving its superiority and availability.https://ieeexplore.ieee.org/document/10110973/Successively sampled seismic data reconstructiondeep learningU-Netfully connected network |
| spellingShingle | Shengjun Li Jianhu Gao Jinyong Gui Lukun Wu Naihao Liu Dongyang He Xin Guo Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data IEEE Access Successively sampled seismic data reconstruction deep learning U-Net fully connected network |
| title | Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data |
| title_full | Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data |
| title_fullStr | Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data |
| title_full_unstemmed | Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data |
| title_short | Fully Connected U-Net and Its Application on Reconstructing Successively Sampled Seismic Data |
| title_sort | fully connected u net and its application on reconstructing successively sampled seismic data |
| topic | Successively sampled seismic data reconstruction deep learning U-Net fully connected network |
| url | https://ieeexplore.ieee.org/document/10110973/ |
| work_keys_str_mv | AT shengjunli fullyconnectedunetanditsapplicationonreconstructingsuccessivelysampledseismicdata AT jianhugao fullyconnectedunetanditsapplicationonreconstructingsuccessivelysampledseismicdata AT jinyonggui fullyconnectedunetanditsapplicationonreconstructingsuccessivelysampledseismicdata AT lukunwu fullyconnectedunetanditsapplicationonreconstructingsuccessivelysampledseismicdata AT naihaoliu fullyconnectedunetanditsapplicationonreconstructingsuccessivelysampledseismicdata AT dongyanghe fullyconnectedunetanditsapplicationonreconstructingsuccessivelysampledseismicdata AT xinguo fullyconnectedunetanditsapplicationonreconstructingsuccessivelysampledseismicdata |