Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial Network
Because of its benefits in terms of high speed, non-destructiveness, and three-dimensionality, as well as ease of integration with computer simulation, computed tomography (CT) technology is widely applied in reservoir geology research. However, rock imaging is restricted by the device used as there...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-07-01
|
Series: | Energies |
Subjects: | |
Online Access: | https://www.mdpi.com/1996-1073/15/14/5115 |
_version_ | 1797406551712989184 |
---|---|
author | Liqun Shan Chengqian Liu Yanchang Liu Weifang Kong Xiali Hei |
author_facet | Liqun Shan Chengqian Liu Yanchang Liu Weifang Kong Xiali Hei |
author_sort | Liqun Shan |
collection | DOAJ |
description | Because of its benefits in terms of high speed, non-destructiveness, and three-dimensionality, as well as ease of integration with computer simulation, computed tomography (CT) technology is widely applied in reservoir geology research. However, rock imaging is restricted by the device used as there is not a win–win for both the image receptive field and corresponding resolution. Convolutional neural network-based super-resolution reconstruction has become a hot topic in improving the performance of CT images. With the help of a convolution kernel, it can effectively extract characteristics and ignore disturbance information. The dismal truth is that convolutional neural networks still have numerous issues, particularly unclear texture details. To address these challenges, a generative adversarial network (RDCA-SRGAN) was designed to improve rock CT image resolution using the combination of residual learning and a dual-channel attention mechanism. Specifically, our generator employs residual attention to extract additional features; similarly, the discriminator builds on dual-channel attention and residual learning to distinguish generated contextual information and decrease computational consumption. Quantitative and qualitative analyses demonstrate that the proposed model is superior to earlier advanced frameworks and is capable to constructure visually indistinguishable high-frequency details. The quantitative analysis shows our model contributes the highest value of structural similarity, enriching the more detailed texture information. From the qualitative analysis, in enlarged details of the reconstructed images, the edges of the images generated by the RDCA-SRGAN can be shown to be clearer and sharper. Our model not only performs well in subtle coal cracks but also enriches more dissolved carbonate and carbon minerals. The RDCA-SRGAN has substantially enhanced the reconstructed image resolution and our model has great potential to be used in geomorphological study and exploration. |
first_indexed | 2024-03-09T03:28:04Z |
format | Article |
id | doaj.art-1779d9a37392467ba5760d59502c81b1 |
institution | Directory Open Access Journal |
issn | 1996-1073 |
language | English |
last_indexed | 2024-03-09T03:28:04Z |
publishDate | 2022-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Energies |
spelling | doaj.art-1779d9a37392467ba5760d59502c81b12023-12-03T14:59:04ZengMDPI AGEnergies1996-10732022-07-011514511510.3390/en15145115Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial NetworkLiqun Shan0Chengqian Liu1Yanchang Liu2Weifang Kong3Xiali Hei4School of Physical and Electrical Engineering, Northeast Petroleum University, Daqing 163318, ChinaSchool of Physical and Electrical Engineering, Northeast Petroleum University, Daqing 163318, ChinaSchool of Physical and Electrical Engineering, Northeast Petroleum University, Daqing 163318, ChinaSchool of Physical and Electrical Engineering, Northeast Petroleum University, Daqing 163318, ChinaSchool of Computing and Informatics, University of Louisiana at Lafayette, Lafayette, LA 70503, USABecause of its benefits in terms of high speed, non-destructiveness, and three-dimensionality, as well as ease of integration with computer simulation, computed tomography (CT) technology is widely applied in reservoir geology research. However, rock imaging is restricted by the device used as there is not a win–win for both the image receptive field and corresponding resolution. Convolutional neural network-based super-resolution reconstruction has become a hot topic in improving the performance of CT images. With the help of a convolution kernel, it can effectively extract characteristics and ignore disturbance information. The dismal truth is that convolutional neural networks still have numerous issues, particularly unclear texture details. To address these challenges, a generative adversarial network (RDCA-SRGAN) was designed to improve rock CT image resolution using the combination of residual learning and a dual-channel attention mechanism. Specifically, our generator employs residual attention to extract additional features; similarly, the discriminator builds on dual-channel attention and residual learning to distinguish generated contextual information and decrease computational consumption. Quantitative and qualitative analyses demonstrate that the proposed model is superior to earlier advanced frameworks and is capable to constructure visually indistinguishable high-frequency details. The quantitative analysis shows our model contributes the highest value of structural similarity, enriching the more detailed texture information. From the qualitative analysis, in enlarged details of the reconstructed images, the edges of the images generated by the RDCA-SRGAN can be shown to be clearer and sharper. Our model not only performs well in subtle coal cracks but also enriches more dissolved carbonate and carbon minerals. The RDCA-SRGAN has substantially enhanced the reconstructed image resolution and our model has great potential to be used in geomorphological study and exploration.https://www.mdpi.com/1996-1073/15/14/5115rock CT imagessuper-resolutionconvolutional neural networksresidual learninggenerative adversarial networkchannel attention mechanism |
spellingShingle | Liqun Shan Chengqian Liu Yanchang Liu Weifang Kong Xiali Hei Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial Network Energies rock CT images super-resolution convolutional neural networks residual learning generative adversarial network channel attention mechanism |
title | Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial Network |
title_full | Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial Network |
title_fullStr | Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial Network |
title_full_unstemmed | Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial Network |
title_short | Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial Network |
title_sort | rock ct image super resolution using residual dual channel attention generative adversarial network |
topic | rock CT images super-resolution convolutional neural networks residual learning generative adversarial network channel attention mechanism |
url | https://www.mdpi.com/1996-1073/15/14/5115 |
work_keys_str_mv | AT liqunshan rockctimagesuperresolutionusingresidualdualchannelattentiongenerativeadversarialnetwork AT chengqianliu rockctimagesuperresolutionusingresidualdualchannelattentiongenerativeadversarialnetwork AT yanchangliu rockctimagesuperresolutionusingresidualdualchannelattentiongenerativeadversarialnetwork AT weifangkong rockctimagesuperresolutionusingresidualdualchannelattentiongenerativeadversarialnetwork AT xialihei rockctimagesuperresolutionusingresidualdualchannelattentiongenerativeadversarialnetwork |