基于双分支残差结构的岩石图像超分辨研究
Research on Rock Image Super-Resolution Based on DBResNet
摘要: 岩石显微图像包含着油气藏分布信息，清晰度高的岩石显微图像包含更多的岩石物理特性。受限于图像采集设备及自然拍摄环境，岩石图像往往存在分辨率低，图像细节不清晰等问题。本文将基于深度学习的卷积神经网络模型应用到岩石图像的超分辨率处理。在对现有网络模型改进的基础上，提出了具有双分支残差结构的超分辨率重建模型，并将改进的模型应用于岩石图像数据集，实现了图像超分辨率效果的改善。实验结果表明本文提出的算法在岩石图像重建的峰值信噪比(PSNR)指标方面有所改善。
Abstract: Rock micro-image contains the information on oil and gas reservoir distribution. The higher rock micro-image resolution is, the more rock physical properties information it contains. Limited by image acquisition equipment and natural shooting environment, rock micro-image often has problems of low resolution and unclear image details. In this paper, deep learning-based convolutional neural network is researched for rock image super-resolution processing. A super-resolution reconstruction model with double branch residual network structure (DBResNet) is proposed after a modification of the current network model. With the application of proposed method to rock image, better super-resolution performance is obtained. The results show that the proposed algorithm improves the peak signal-to-noise ratio (PSNR) of rock image reconstruction.
文章引用：田茹梦, 朱联祥. 基于双分支残差结构的岩石图像超分辨研究[J]. 计算机科学与应用, 2022, 12(2): 338-346. https://doi.org/10.12677/CSA.2022.122034

参考文献

[1] 程国建, 张福临. 基于SinGAN的岩石薄片图像超分辨率重建[J]. 西安石油大学学报(自然科学版), 2021, 36(2): 116-121.
[2] 张德, 林青宇, 郭茂祖. 单幅图像超分辨重建的深度学习方法综述[J]. 计算机工程与应用, 57(22): 1-16.
[3] 李彬, 王平, 赵思逸. 基于双重注意力机制的图像超分辨重建算法[J]. 图学学报, 2021, 42(2): 206-215.
[4] Lim, B., Son, S., Kim, H., Nah, S. and Lee, K.M. (2017) Enhanced Deep Residual Networks for Single Image Super-Resolution. 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Honolulu, 21-26 July 2017, 1132-1140. [Google Scholar] [CrossRef] 
[5] Dong, C., Loy, C.C., He, K.M. and Tang, X.O. (2014) Learning a Deep Convolutional Network for Imagesuper-Resolution. European Conference on Computer Vision (ECCV), Zurich, September 2014, 184-199. [Google Scholar] [CrossRef] 
[6] Kim, J., Lee, J.K. and Lee, K.M. (2015) Accurate Image Su-per-Resolution Using Very Deep Convolutional Networks. IEEE Conference on Computer Vision and Pattern Recogni-tion (CVPR), LasVegas, July 2016, 1646-1654..
[7] Ledig, C., Theis, L., Huszar, F., Caballero, J., Aitken, A., Tejani, A., Totz, J., Wang, Z.H. and Shi, W.Z. (2017) Photo-Realistic Single Image Super-Resolution Using a Generative Ad-versarial Network. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, 21-26 July 2017, 105-114. [Google Scholar] [CrossRef] 
[8] Dong, C., Loy, C.C. and Tang, X.O. (2016) Accelerating the Super-Resolution Convolutional Neural Network. Proceedings of the European Conference on Computer Vision (ECCV), Amsterdam, October 2016, 391-407. [Google Scholar] [CrossRef] 
[9] Tong, T., Li, G., Liu, X.J. and Gao, Q.Q. (2017) Image Su-per-Resolution Using Dense Skip Connections. 2017 IEEE International Conference on Computer Vision (ICCV), Ven-ice, 22-29 October 2017, 4809-4817. [Google Scholar] [CrossRef] 
[10] Zhang, Y.L., Tian, Y.P., Kong, Y., Zhong, B.N. and Fu, Y. (2018) Residual Dense Network for Image Super-Resolution. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Lake City, 18-23 June 2018, 2472-2481. [Google Scholar] [CrossRef] 
[11] 余晓露, 叶恺, 杜崇娇, 宫晗凝, 马中良. 基于卷积神经网络的碳酸盐岩生物化石显微图像识别[J]. 石油实验地质, 2021, 43(5): 880-885+895.
[12] 李现国, 冯欣欣, 李建雄. 多尺度残差网络的单幅图像超分辨率重建[J]. 计算机工程与应用, 2021, 57(7): 215-221.