自注意力SRGAN在岩石CT图像超分辨中的应用研究Applications of Self-attention SRGAN in Super Resolution Reconstruction of Rock CT Image
朱联祥;郑逸;
摘要(Abstract)：
针对图像采集设备和地质环境等因素导致的岩石显微图像普遍存在分辨率低，图像细节不清晰等问题，本文基于SRGAN和自注意力机制，对生成网络和判别网络进行调整，并在损失函数中结合岩石显微图像的孔隙度特征，提出一种岩石显微图像超分辨率重建算法。在4倍放大因子下，通过实验对岩石显微图像数据集DRSRD1＿2D进行了测试。结果表明，本文所提算法在重建图像的峰值信噪比(PSNR)指标方面有显著改善。与SRGAN算法相比，在运行相同轮数的情况下，重建结果的边缘细节更清晰且亮度信息更准确，能更好地表达图像的高频特征。
关键词(KeyWords)： 岩石显微图像;超分辨率重建;生成对抗网络;自注意力机制;孔隙度
基金项目(Foundation): 移动通信教育部工程研究中心开放研究项目(cqupt-mct-202006)
作者(Authors): 朱联祥;郑逸;
参考文献(References)：
- [1] 张虹，董国臣，王松洋，等.浅析显微CT技术在油气储层研究中的应用[J].资源与产业，2013,15(6):98-102.ZHANG Hong,DONG Guochen,WANG Songyang,et al.Application of microscopic CT in oil-gas reservoir study[J].Resources & Industries.2013,15(6):98-102.
- [2] ANWAR S,KHAN S,BARNES N.A deep journey into super-resolution:A survey[J].ACM Computing Surveys (CSUR),2020,53(3):1-34.
- [3] DONG C,LOY C C,HE K,et al.Learning a deep convolutional network for image super-resolution[C]//European Conference on Computer Vision.Springer,Cham,2014:184-199.
- [4] SHI W,CABALLERO J,Huszár F,et al.Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:1874-1883.
- [5] HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:770-778.
- [6] LAI W S,HUANG J B,AHUJA N,et al.Deep laplacian pyramid networks for fast and accurate super-resolution[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:624-632.
- [7] GOODFELLOW I J,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial networks[J].Advances in Neural Information Processing System,2014,27.
- [8] LEDIG C,THEIS L,Huszár F,et al.Photo-realistic single image super-resolution using a generative adversarial network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:4681-4690.
- [9] BAHDANAU D,CHO K,BENGIO Y.Neural machine translation by jointly learning to align and translate[C]//arXiv preprint arXiv:1409.0473,2014.
- [10] VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need[J].Advances in neural information processing system,2017,30.
- [11] 陈超，魏彪，梁婷，等.一种基于工业CT技术的岩芯样品孔隙度测量分析方法[J].物探与化探，2013(3):500-507.CHEN Chao,WEI Biao,LIANG Ting,et al.The application of industrial computation tomography(CT) to the analysis of core sample porosity[J].Geophysical and Geochemical Exploration,2013(3):500-507.