基于循环一致性对抗网络的稀疏阵列宽带合成波束效应消除
Wideband Beam Effect Elimination of Sparse Array Based on Cyclic Consistency Adversarial Network
DOI: 10.12677/ORF.2023.136751, PDF,    国家自然科学基金支持
作者: 周 娟, 罗珺茜, 彭 嘉, 龙楷潮, 张 利*：贵州大学大数据与信息工程学院，贵州 贵阳
关键词: 深度学习；图像重建；宽带合成波束；射电干涉；Deep Learning； Image Reconstruction； Wideband Beam； Radio Interference
摘要: 射电天文稀疏干涉阵列成像过程中由于无线电接收器的带宽限制会使观测后的图像产生宽带合成波束效应，针对这个问题，本文设计了一个带宽合成波束效应的消除CycleGAN模型。该模型利用CycleGAN模型中的残差学习机制对具有空间复杂结构的射电源信号携带的带宽合成波束效应图像进行特征提取，进而提高恢复效果。通过天文通用软件CASA模拟出来的脏图和原图作为图像对模型进行训练。这种结合方式能够有效地将两种图像风格进行转换，从而使得模型能够更好地适应不同的射电源信号。实验结果显示，该深度学习算法与现有的宽带合成波束算法在图像指标PSNR和SSIM上得到明显提升，能够有效地恢复天空图像，这一技术将为我们提供更为准确的天文学数据，并推动天文学的发展。
Abstract: Due to the bandwidth limitation of the radio receiver, wideband beam effect is generated in the imaging of radio astronomy sparse interference array. To solve this problem, a CycleGAN model is designed to eliminate the beam effect. In this model, the residual learning mechanism of CycleGAN model is used to extract the features of bandwidth-synthesised beame ffect images car-ried by radio source signals with complex spatial structure, so as to improve the recovery effect. The model was trained by using the dirty map and the original image simulated by CASA—a astronomical software. This combination method can effectively convert the two image styles, so that the model can better adapt to different radio source signals. The experimental results show that the deep learning algorithm is significantly improved in PSNR and SSIM compared with the existing wideband synthetic beam algorithm, and can effectively restore the sky image. This technology will provide us with more accurate astronomical data and promote the development of astronomy.
文章引用：周娟, 罗珺茜, 彭嘉, 龙楷潮, 张利. 基于循环一致性对抗网络的稀疏阵列宽带合成波束效应消除[J]. 运筹与模糊学, 2023, 13(6): 7663-7673. https://doi.org/10.12677/ORF.2023.136751

参考文献

[1] Ogod, V.M., Gel’Freikh, G.B., Willson, R.F., et al. (1992) Very Large Array. Solar Physics, 141, 303-323. [Google Scholar] [CrossRef] 
[2] Napier, P.J., Thompson, A.R. and Ekers, R.D. (1983) The Very Large Array: Design and Performance of Amodern Synthesis Radio Telescope. Proceedings of the IEEE, 71, 1295-1320. [Google Scholar] [CrossRef] 
[3] Thompson, A.R., Moran, J.M. and Swenaon, G.W. (2017) Interferometry and Synthesis in Radio Astronomy. 3rd Edition, Springer International Publishing, Berlin. [Google Scholar] [CrossRef] 
[4] Venkata, U.R. (2010) Parameterized Deconvolution for Wide-Band Radio Synthesis Imaging. Ph.D. Thesis, New Mexico Institute of Mining and Technology Socorro, New Mexico.
[5] 卫星奇, 张利, 吴康宁, 卢梅, 王蓓, 贺春林, 潘伟. 基于深度学习的低频SKA带宽涂污效应矫正方法[J]. 软件工程与应用, 2022, 11(1): 72-80. [Google Scholar] [CrossRef] 
[6] Högbom, J.A. (1974) Aperture Synthesis with a Non-Regular Distribution of Inter⁃Ferometer Baseline. Astronomy and Astrophysics Supplement Series, 15, 417-426.
[7] Clark, B.G. (1980) An Efficient Implementation of the Algorithm “CLEAN”. Astronomy & Astrophysics, 89, 377-378.
[8] Cornwell, T.J. (2008) Multiscale CLEAN Deconvolution of Radio Synthesis Images. IEEE Journal of Selected Topics in Signal Processing, 25, 793-801. [Google Scholar] [CrossRef] 
[9] Zhang, L., Bhatnagar, S., Rau, U. and Zhang, M. (2016) Efficient Implementation of the Adaptive Scale Pixel Decompositioin Algorithm. Astronomy & Astrophysics, 592, A128. [Google Scholar] [CrossRef] 
[10] Zhang, L., Xu, L., Zhang, M. and Wu, Z.Z. (2019) An Adaptive Loop Gain Selection for CLEAN Deconvolution Algorithm. Research in Astronomy and Astrophysics, 19, Article No. 79. [Google Scholar] [CrossRef] 
[11] Zhang, L., Mi, L.G., Zhang, M., Liu, X. and He, C.L. (2020) Adaptive-Scale Wide-Field Reconstruction for Radio Synthesis Imaging. Astronomy & Astrophysics, 640, A80. [Google Scholar] [CrossRef] 
[12] Zhang, L., Mi, L.G., Xu, L., Zhang, M., Li, D.Y., Liu, X., Wang, F., Xiao, Y.F. and Wu, Z.Z. (2021) Adaptive Scale Model Reconstruction for Radio Synthesis Imaging. Research in Astronomy and Astrophysics, 21, Article No. 63. [Google Scholar] [CrossRef] 
[13] Zhang, L., Mi, L.G., Zhang, M., Liu, X., Xu, L., Wang, F., Ruan, Y.J. and Li, D.Y. (2021) Parameterized Reconstruction with Random Scales for Radio Synthesis Imaging. Astronomy & Astrophysics, 646, A44. [Google Scholar] [CrossRef] 
[14] Goodfellow, I., Pouget-Abadie, J., et al. (2020) Genera-tive Adversarial Networks. Communications of the ACM, 63, 139-144. [Google Scholar] [CrossRef] 
[15] Denton, E.L., Chintala, S., Fergus, R., et al. (2015) Deep Generative Image Models Using a Laplacian Pyramid of Adversarial Networks.
[16] Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A. and Chen, X. (2016) Improved Techniques for Training GANs.
[17] Wang, T.C., Liu, M.Y., Zhu, J.Y., et al. (2018) High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, 18-23 June 2018, 8798-8807. [Google Scholar] [CrossRef] 
[18] Fang, W., Zhang, F., Sheng, V.S., et al. (2018) A Method for Improving CNN-Based Image Recognition Using DCGAN. Computers, Materials & Continua, 57, 167-178. [Google Scholar] [CrossRef] 
[19] Brock, A., Donahue, J. and Simonyan, K. (2018) Large Scale GAN Training for High Fidelity Natural Image Synthesis.
[20] Alotaibi, A. (2020) Deep Generative Adversarial Networks for Image-to-Image Translation: A Review. Symmetry, 12, Article No. 1705. [Google Scholar] [CrossRef] 
[21] Ji, W., Guo, J. and Li, Y. (2020) Multi‐Head Mutual-Attention Cyclegan for Unpaired Image-to-Image Translation. IET Image Processing, 14, 2395-2402. [Google Scholar] [CrossRef] 
[22] Johnson, J., Alahi, A. and Li, F.-F. (2016) Perceptual Losses for Real-Time Style Transfer and Super-Resolution. 14th European Conference, Computer Vision—ECCV 2016, Am-sterdam, 11-14 October 2016, 694-711.
[23] He, K., Zhang, X., Ren, S. and Sun, J. (2016) Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 27-30 June 2016, 770-778. [Google Scholar] [CrossRef] 
[24] Shrivastava, A., Pfister, T., Tuzel, O., Susskind, J., Wang, W. and Webb, R. (2017) Learning from Simulated and Unsupervised Images through Adversarial Training. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, 21-26 July 2017, 2242-2251. [Google Scholar] [CrossRef] 
[25] Goodfellow, I. (2016) NIPS 2016 Tutorial: Generative Adver-sarial Networks.
[26] Alam, S., Albareti, F.D., Prieto, C.A., et al. (2015) The Eleventh and Twelfth Data Releases of the Sloan Digital Sky Survey: Final Data from SDSS-III. The Astrophysical Journal Supplement Series, 219, 12-39. [Google Scholar] [CrossRef] 

为你推荐