基于CF-PSPnet的主波束效应的校正研究
Correction Study for Primary Beam Effect Based on CF-PSPnet
DOI: 10.12677/ORF.2023.136758, PDF,    国家自然科学基金支持
作者: 马 帅, 陈 港, 韦焕泽, 雷 然, 张 利*：贵州大学大数据与信息工程学院，贵州 贵阳
关键词: 射电天文成像；深度学习；主波束效应；主波束校正；Radio Astronomical Imaging； Deep Learning； Primary Beam Effect； Primary Beam Correction
摘要: 射电天文成像过程中，需要对观测得到的脏图进行反卷积操作得到“干净”的图像，而这图像中包含着主波束效应，未得到主波束效应校正的图像在偏离中心区域的地方会出现强度衰减的现象，而且距离中心越远，衰减现象越明显。传统的主波束校正方法虽然能够完成校正，但对于图像边缘区域尤其是微弱结构的细节恢复的不是很理想。因此，本文提出了CF-PSPnet的深度学习的模型，引入了注意力机制，多特征融合等操作，实现了对观测图像中不同级别特征的融合与提取，从而达到对主波束校正的需求。经本文实验后得出，CF-PSPnet解决了传统图像域校正方法中存在的缺点，有着不错的校正效果。下一代的射电天文望远镜有着更高灵敏度，动态范围与空间角分辨率，本文方法为其主波束效应校正提供了一种更具潜力的解决方案。
Abstract: In the process of radio astronomy imaging, it is necessary to deconvolve the dirty image obtained from observation to obtain a “clean” image, which contains the primary beam effect, and the image not corrected for the primary beam effect will have intensity attenuation in the off-centre area, and the further away from the centre, the more obvious the attenuation phenomenon. Although the traditional primary beam correction method can complete the correction, it is not ideal for the image edge region, especially for the recovery of the details of the weak structure. Therefore, this paper proposes a deep learning model of CF-PSPnet, which introduces the attention mechanism, multi-feature fusion and other operations to achieve the fusion and extraction of different levels of features in the observed image, so as to achieve the demand for the primary beam correction. After the experiments in this paper, it is concluded that CF-PSPnet solves the shortcomings of the traditional image domain correction method and has good correction effect. The next generation of radio astronomy telescopes has higher sensitivity, dynamic range and spatial angular resolution, and the method in this paper provides a more promising solution for its main beam effect correction.
文章引用：马帅, 陈港, 韦焕泽, 雷然, 张利. 基于CF-PSPnet的主波束效应的校正研究[J]. 运筹与模糊学, 2023, 13(6): 7757-7767. https://doi.org/10.12677/ORF.2023.136758

参考文献

[1] Bhatnagar, S., Cornwell, T.J., Golap, K., et al. (2008) Correcting Direction-Dependent Gains in the Deconvolution of Radio Interferometric Images. Astronomy & Astrophysics, 487, 419-429. [Google Scholar] [CrossRef] 
[2] Thompson, A.R. (1999) Fundamentals of Radio Interfer-ometry. Synthesis Imaging in Radio Astronomy II, 180, 11.
[3] Cheng, J. (2009) The Principles of Astronomical Telescope Design. Springer, New York. [Google Scholar] [CrossRef] 
[4] Pearson, T.J. and Readhead, A.C.S. (1984) Image Formation by Self-Calibration in Radio Astronomy. Annual Review of Astronomy and Astrophysics, 22, 97-130. [Google Scholar] [CrossRef] 
[5] Tanoglidis, D., Ciprijanovic, A. and Drli-ca-Wagner, A. (2021) Deep Shadows: Separating Low Surface Brightness Galaxies from Artifacts Using Deep Learning. Astronomy and Computing, 35, 100469. [Google Scholar] [CrossRef] 
[6] Czech, D., Mishra, A. and Inggs, M. (2018) A CNN and LSTM-Based Approach to Classifying Transient Radio Frequency Interference. Astronomy and Computing, 25, 52-57. [Google Scholar] [CrossRef] 
[7] Cabrera-Vives, G., Reyes, I., Forster, F., Estrvez, P.A. and Maureira, J.C. (2017) Deep-Hits: Rotation Invariant Convolutional Neural Network for Transient Detection. arXiv:1701.00458. [Google Scholar] [CrossRef] 
[8] Meher, S.K. and Panda, G. (2021) Deep Learning in Astronomy: A Tutorial Perspective. The European Physical Journal Special Topics, 230, 2285-2317. [Google Scholar] [CrossRef] 
[9] Hamaker, J.P., Bregman, J.D. and Sault, R.J. (1996) Understanding Radio Polarimetry. I. Mathematical Foundations. Astronomy and Astrophysics Supplement Series, 117, 137-147. [Google Scholar] [CrossRef] 
[10] Swenson Jr, G.W. (1969) Synthetic-Aperture Radio Telescopes. Annual Review of Astronomy and Astrophysics, 7, 353. [Google Scholar] [CrossRef] 
[11] Venkata, U.R. (2010) Parameterized Deconvolu-tion for Wide-Band Radio Synthesis Imaging. PhD Thesis, New Mexico Institute of Mining and Technology, 2010, 281.
[12] Zhang, L., Xu, L. and Zhang, M. (2020) Parameterized Clean Deconvolution in Radio Synthesis Imaging. Publications of the Astronomical Society of the Pacific, 132, 041001. [Google Scholar] [CrossRef] 
[13] Zhao, H., Shi, J., Qi, X., et al. (2017) Pyramid Scene Parsing Network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 2881-2890. [Google Scholar] [CrossRef] 
[14] Wang, P., Chen, P., Yuan, Y., et al. (2018) Understanding Convolution for Semantic Segmentation. 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), 12-15 March 2018, Tahoe, 1451-1460. [Google Scholar] [CrossRef] 
[15] Bean, B., Bhatnagar, S., Castro, S., et al. (2022) CASA, Common Astronomy Software Applications for Radio Astronomy. Publications of the Astronomical Society of the Pacific, 134, 114501. [Google Scholar] [CrossRef] 
[16] Horé, A. and Ziou, D. (2010) Image Quality Metrics: PSNR vs. SSIM. 20th International Conference on Pattern Recognition, ICPR 2010, Istanbul, 23-26 August 2010, 2366-2369. [Google Scholar] [CrossRef] 
[17] Zhou, W., Bovik, A.C., Sheikh, H.R., et al. (2004) Image Quality Assessment: From Error Visibility to Structural Similarity. IEEE Transactions on Image Processing, 13, 600-612. [Google Scholar] [CrossRef] 

为你推荐