基于对偶循环结构及注意力机制的红外图像超分辨率重建
Super Resolution Reconstruction of Infrared Image Based on Dual Cycle Structure and Attention Mechanism
DOI: 10.12677/CSA.2022.124081, PDF,    国家自然科学基金支持
作者: 闫昊天, 程良伦, 吴 衡：广东工业大学，自动化学院，广东 广州
关键词: 红外图像；超分辨率重建；对偶循环结构；注意力机制；Infrared Image； Super-Resolution Reconstruction； Dual Loop Structure； Attention Mechanism
摘要: 针对红外图像空间分辨率低、成像效果不好的问题，提出了一种基于对偶循环结构和注意力机制的红外图像超分辨率重建方法。对偶循环结构的引入能够更好地约束LR到HR的映射，通过引入融合了多维度的注意力机制CBAM，让网络在不显著提高计算量与参数的情况下取得了细节更丰富的重建效果。通过在真实红外数据集上与现有的典型方法进行比较，所提方法在显著降低了模型的参数量的情况下取得了不错的重建效果。
Abstract: Aiming at the problems of low spatial resolution and poor imaging effect of infrared image, a super-resolution reconstruction method of infrared image based on dual cycle structure and attention mechanism is proposed. The introduction of dual loop structure can better restrict the mapping from LR to HR. By introducing CBAM, which integrates multi-dimensional attention mechanism, the network can achieve more detailed reconstruction without significantly improving the amount of calculation and parameters. Compared with several existing typical methods, the proposed method achieves good results under the condition of significantly reducing the number of parameters of the model.
文章引用：闫昊天, 程良伦, 吴衡. 基于对偶循环结构及注意力机制的红外图像超分辨率重建[J]. 计算机科学与应用, 2022, 12(4): 797-805. https://doi.org/10.12677/CSA.2022.124081

参考文献

[1] 王岭雪, 蔡毅. 红外成像光学系统进展与展望[J]. 红外技术, 2019, 41(1): 1-12.
[2] 莫朝霞, 陈沅江. 我国红外热像检测技术的研究及发展展望[J]. 激光与红外, 2014, 44(12): 1300-1305.
[3] 汪金礼, 王皓, 袁宝红, 刘鸿涛, 苏亚辉, 尹涛. 红外视频监控中的人体目标检测[J]. 测试技术学报, 2014, 28(1): 39-43.
[4] Ragb, H. and Asari, V. (2019) Multifeature Fusion for Robust Human Detection in Thermal Infrared Imagery. Optical Engineering, 58, Arti-cle ID: 043101. [Google Scholar] [CrossRef] 
[5] Zhou, Z., et al. (2016) Fusion of Infrared and Visi-ble Images for Night-Vision Context Enhancement. Applied Optics, 55, 6480-6490. [Google Scholar] [CrossRef] 
[6] Zafar, I., et al. (2010) Human Silhouette Extraction on FPGAs for In-frared Night Vision Military Surveillance. 2010 Second Pacific-Asia Conference on Circuits, Communications and Sys-tem, Beijing, 1-2 August 2010, 63-66. [Google Scholar] [CrossRef] 
[7] Chen, H.W., et al. (2021) Advanced Automatic Target Recog-nition (ATR) with Infrared (IR) Sensors. 2021 IEEE Aerospace Conference (50100), Big Sky, 6-13 March 2021, 1-13. [Google Scholar] [CrossRef] 
[8] Kulchitsky, N.A., Naumov, A.V. and Startsev, V.V. (2020) Development Trends of Infrared Focal Plane Array Detectors. [Google Scholar] [CrossRef] 
[9] Liangpei, et al. (2010) A Super-Resolution Reconstruction Algorithm for Surveillance Images. Signal Processing, 90, 848-859. [Google Scholar] [CrossRef] 
[10] Keys, R.G. (1982) Cubic Convolution Interpolation for Digital Image Processing. IEEE Transactions on Acoustics, Speech, and Signal Processing, 29, 1153-1160. [Google Scholar] [CrossRef] 
[11] Thurnhofer and Stefan (1996) Edge-Enhanced Image Zooming. Optical Engineering, 35, 1862-1870. [Google Scholar] [CrossRef] 
[12] Dong, C., et al. (2014) Learning a Deep Convolutional Network for Image Super-Resolution. 13th European Conference, Zurich, 6-12 September 2014, 184-199. [Google Scholar] [CrossRef] 
[13] Kim, J., Lee, J.K. and Lee, K.M. (2016) Accurate Image Su-per-Resolution Using Very Deep Convolutional Networks. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 27-30 June 2016, 1646-1654. [Google Scholar] [CrossRef] 
[14] Ledig, C., et al. (2017) Photo-Realistic Single Image Su-per-Resolution Using a Generative Adversarial Network. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, 21-26 July 2017, 105-114. [Google Scholar] [CrossRef] 
[15] Wang, X., et al. (2018) Esrgan: Enhanced Super-Resolution Genera-tive Adversarial Networks. 15th European Conference on Computer Vision, Munich, 8-14 September 2018, 1-16.
[16] Zhang, Y., et al. (2018) Image Super-Resolution Using Very Deep Residual Channel Attention Networks. 15th European Conference on Computer Vision, Munich, 8-14 September 2018, 294-310. [Google Scholar] [CrossRef] 
[17] Ronneberger, O., Fischer, P. and Brox, T. (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. Springer International Publishing, Berlin. [Google Scholar] [CrossRef] 
[18] Guo, Y., et al. (2020) Closed-Loop Matters: Dual Regression Networks for Single Image Super-Resolution. IEEE Conference on Computer Vision and Pattern Recognition, Seattle, 13-19 June 2020, 5406-5415. [Google Scholar] [CrossRef] 
[19] Woo, S., Park, J., Lee, J.Y., et al. (2018) CBAM: Convolu-tional Block Attention Module. Proceedings of the European Conference on Computer Vision (ECCV), Munich, 8-14 September 2018, 3-19. [Google Scholar] [CrossRef]