|
|
|
|
|
|
| The “Unknown” Spectral Classification Study of LAMOST:
ODS-YOLOv7 Model |
| WANG Xiao-min1, GAO Jun-ping1*, PU Yuan2*, QIU Bo1*, ZHANG Jian-nan3, YAN Jing1, LI Rong1 |
1. Hebei University of Technology, Tianjin 300400，China
2. Guangdong Baiyun University, Guangzhou 510450，China
3. National Astronomical Observatories, Chinese Academy of Sciences, Beijing 100012, China
|
|
|
|
|
Abstract Identifying celestial spectra is essential for making new astronomical discoveries and conducting detailed studies of celestial objects. The LAMOST DR8 v1.0 release of low-resolution spectral data contains approximately 530 000 spectra named “Unknown”. The reason is that they have no category labels. And 88.56% of these spectra have signal-to-noise ratios between 0 and 10. Therefore, the effective output of LAMOST will increase if we analyze these spectra. In this paper, we propose an ODS-YOLOv7 model to deal with the problem of the “Unknown” spectral classification. It is an end-to-end category prediction model and is suitable for one-dimensional spectra. We also add a one-dimensional convolutional attention module to improve the accuracy of spectra recognition. After training on a set of known category spectra with signal-to-noise ratios between 0 and 10, the ODS-YOLOv7 model can learn the effective features of the low signal-to-noise spectra. Thus, it can enable us to predict “Unknown” spectra. Experiments show that the model has an F1-score of 0.98, 0.95, and 0.95 for the spectral identification of low signal-to-noise stars, galaxies, and quasars spectra with known labels. In the meantime, ODS-YOLOv7 obtains the best results in comparison experiments with traditional algorithms KNN, RF, DT, SVM, and deep learning algorithms 1D CNN, 1DSSCNN, ResNet, DenseNet, and VIT. The experimental results also give confidence in the predictions of the ODS-YOLOv7 model for the “Unknown” spectra in DR8 v1.0, with 92% of the confidence levels above 60%. To ensure the quality of the model output, only spectral categories with a prediction confidence level greater than 99% are selected as output in this paper. Ultimately, 37.19% and 47.03% of the “Unknown” spectra released in DR8 v1.0 and DR9 v0, respectively, are predicted by this model.In addition, the paper tests the accuracy of the model's predictions using manual authentication. To improve the interpretability of the model,the paper takes the Grad-CAM method for two-dimensional image visualisation. It improves it into an algorithm suitable for visualising one-dimensional spectral data to predict output features. Experiments show that the model focuses on different features in the visualisation of different classes of astronomical features and that the model is good at predicting low signal-to-noise “unknown” spectral classes.
|
|
Received: 2023-05-09
Accepted: 2023-09-25
|
|
|
|
Corresponding Authors:
GAO Jun-ping, PU Yuan, QIU Bo
E-mail: gaopingcn@126.com；puyuan@baiyunu.edu.cn；qiubo@hebut.edu.cn
|
|
[1] Corral L J, Fierro-Santillán C R. arXiv preprint arXiv: 2105. 07110, 2021.
[2] Wang L L, Shen S Y, Luo A L, et al. The Astrophysical Journal Supplement Series, 2022, 258(1): 9.
[3] Weaver W B, Torres-Dodgen A V. The Astrophysical Journal, 1995, 446: 300.
[4] Sharma K, Kembhavi A, Kembhavi A, et al. Monthly Notices of the Royal Astronomical Society, 2020, 491(2): 2280.
[5] Wen Xiaoqing, Yang Jinmeng. Chinese Journal of Physics, 2021, 69: 303.
[6] HE Dong-yuan, LIU Wei, CAO Shuo, et al(何东远, 刘 伟, 曹 硕, 等). Journal of Beijing Normal University(北京师范大学学报), 2020, 56(1): 37.
[7] WANG Qi-xun, ZHAO Gang, FAN Zhou(王奇勋, 赵 刚, 范 舟). Astronomical Research and Technology(天文研究与技术), 2020, 17(1): 85.
[8] HONG Shu-xin, ZOU Zhi-qiang, XU Ling-zhe(洪舒欣, 邹志强, 徐灵哲). Acta Astronomica Sinica(天文学报), 2021, 62(5): 48.
[9] Zheng Z P, Qiu B, Luo A L, et al. Publications of the Astronomical Society of the Pacific, 2020, 132(1008): 024504.
[10] Guo Y X, Luo A L, Zhang S, et al. Monthly Notices of the Royal Astronomical Society, 2019, 485(2): 2167.
[11] YANG Yu-qing, CAI Jiang-hui, YANG Hai-feng, et al(杨雨晴, 蔡江辉, 杨海峰, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2022, 42(4): 1186.
[12] Li X R, Lin Y T, Qiu K B. Research in Astronomy and Astrophysics, 2019, 19(8): 111.
[13] He K, Zhang X, Ren S, et al. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016: 770.
[14] Dosovitskiy A, Beyer L, Kolesnikov A, et al. arXiv preprint arXiv: 2010. 11929, 2020.
[15] Woo S, Park J, Lee J Y, et al. Proceedings of the European Conference on Computer Vision (ECCV). 2018: 3.
[16] Selvaraju R R, Cogswell M, Das A, et al. Proceedings of the IEEE International Conference on Computer Vision，2017: 618.
|
| [1] |
TU Liang-ping1, 2, LI Shuang-chuan2, TU Dong-xin1*, LI Jian-xi*, DING Zhi-chao2. SATS: A Stellar Spectral Classification Algorithm Based on Multiple Feature Extraction[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(07): 2029-2036. |
| [2] |
LI Hao1, ZHAO Qing1*, CUI Chen-zhou2, FAN Dong-wei2, ZHANG Cheng-kui1, SHI Yan-cui1, WANG Yuan1. A Stellar Spectrum Classification Algorithm Based on CNN and LSTM Composite Deep Learning Model[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(06): 1668-1675. |
| [3] |
LI Shuang-chuan, TU Liang-ping*, LI Xin, WANG Li-li. Besvm: A-Type Star Spectral Subtype Classification Algorithm Based on Transformer Feature Extraction[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(05): 1575-1581. |
| [4] |
YANG Yu-qing1, CAI Jiang-hui1, 2*, YANG Hai-feng1*, ZHAO Xu-jun1, YIN Xiao-na1. LAMOST Unknown Spectral Analysis Based on Influence Space and Data Field[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1186-1191. |
| [5] |
MA Yang, ZHANG Ji-fu, CAI Jiang-hui, YANG Hai-feng, ZHAO Xu-jun*. Parallel Extraction and Analysis of Abnormal Features of QSO Spectra Based on Sparse Subspace[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(04): 1086-1091. |
| [6] |
QU Cai-Xia, YANG Hai-feng*, CAI Jiang-hui, XUN Ya-ling. P-Cygni Profile Analysis of the Spectrum: LAMOST J152238.11+333136.1[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2020, 40(04): 1304-1308. |
| [7] |
ZHENG Guo-liang, ZHU Hong-qiu*, LI Yong-gang. Spectral Signal Denoising Algorithm Based on Improved LMS[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2020, 40(02): 643-649. |
| [8] |
WANG Meng-xin 1，2, LUO A-li1，2. Spectral Analysis of Host Galaxy from Possible Dual AGNs[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2020, 40(01): 290-293. |
| [9] |
BU Yu-de1*, PAN Jing-chang2, YI Zhen-ping2. Estimation of Mg Abundances from Low Resolution Spectra Using ELM Algorithm[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2019, 39(10): 3288-3291. |
| [10] |
ZHANG Xiao1，2, LUO A-li1*. XGBOOST Based Stellar Spectral Classification and Quantized Feature[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2019, 39(10): 3292-3296. |
| [11] |
XUE Ren-zheng1, CHEN Shu-xin1*, HUANG Hong-ben2. Line Index of A-Type Stellar Astronomical Spectrum Predict Effective Temperature by Ridge Regression Model[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2019, 39(08): 2624-2629. |
| [12] |
QU Cai-xia1, YANG Hai-feng1*, CAI Jiang-hui1*, LUO A-li2, ZHANG Ji-fu1, NIE Yao-yao1. Feature Extraction and Analysis of Double-Peaked Emission Line Spectra Based on Relevant Subspace[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2019, 39(06): 1677-1682. |
| [13] |
CHEN Shu-xin 1, 2, SUN Wei-min1*, SONG Yi-han3. Study on the Classification of LAMOST Early Stellar Spectrum Template by Mahalanobis Distance[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2019, 39(05): 1618-1622. |
| [14] |
CAI Jiang-hui1, YANG Yu-qing1, YANG Hai-feng1*, LUO A-li2, KONG Xiao2, ZHANG Ji-fu1. Spectral Analysis of Sky Light Based on Trajectory Clustering[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2019, 39(04): 1301-1306. |
| [15] |
WU Ming-lei1, PAN Jing-chang1*, YI Zhen-ping1, WEI Peng2. A Method to Search Special Stellar Spectra from Low Signal-to-Noise Ratio Spectral Sky Survey Data[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2019, 39(02): 618-621. |
|
|
|
|
