|
|
|
|
|
|
| TDSC-Net: A Two-Dimensional Stellar Spectra Classification Model Based on Attention Mechanism and Feature Fusion |
| LI Rong1, CAO Guan-long1*, PU Yuan2*, QIU Bo1, WANG Xiao-min1, YAN Jing1, WANG Kun1 |
1. Hebei University of Technology, Tianjin 300400, China
2. Guangdong Baiyun University, Guangzhou 510450, China
|
|
|
|
|
Abstract Traditional one-dimensional spectra perform poorly when classifying celestial objects with a low signal-to-noise ratio (SNR). Therefore, the paper uses two-dimensional spectra and proposes a feature fusion model called TDSC-Net (Two-Dimensional Spectra Classification Network), incorporating an attention mechanism for stellar classification. TDSC-Net employs identical feature extraction layers to get features from the two-dimensional spectra of stars, specifically from the blue and red ends. The extracted features are fused and employed for the classification task. The stellar spectral data in this experiment is selected from the LAMOST (the Large Sky Area Multi-Object Fiber Spectroscopic Telescope) database. They are using Z-score normalization on the spectra to reduce convergence difficulties caused by significant variations in spectral flux values and evaluating the model performance by four metrics: Precision, Recall, F1-score, and Accuracy. The experiments consist of two parts: In the first part, TDSC-Net is employed to classify A, F, G, K, and M-type stars to validate the reliability of using two-dimensional spectra for multi-class stellar classification. In the second part, the two-dimensional spectra are classified based on different SNRs to investigate the impact of SNRs on classification accuracy. The first part's results show that the five-class classification accuracy reaches 84.3%. The classification accuracies of A, F, G, K, and M types are 87.0%, 84.6%, 81.2%, 87.4%, and 89.7%, respectively. These accuracies are higher than the results obtained from one-dimensional spectra classification after spectra extraction. The results of the second part indicate that even in the low SNR (SNR<30), the accuracy of two-dimensional spectra classification can still reach 78.9%. Once the SNR surpasses 30, the impact of SNR on spectra classification becomes less significant. These provide evidence for the importance of using two-dimensional spectra classification in low SNR and demonstrate the effectiveness of TDSC-Net in stellar spectra classification.
|
|
Received: 2023-05-22
Accepted: 2023-09-18
|
|
|
|
Corresponding Authors:
CAO Guan-long, PU Yuan
E-mail: caoguanlong@hebut.edu.cn；puyuan@baiyunu.edu.cn
|
|
[1] Lu Y, Luo A L, Wang L L, et al. Astronomy and Computing, 2021, 36: 100485.
[2] Kheirdastan S, Bazarghan M. Astrophysics and Space Science, 2016, 361(9): 304.
[3] TU Liang-ping, WEI Hui-ming, WANG Zhi-heng, et al(屠良平，魏会明，王志衡，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2015, 35(4): 1103.
[4] Xing F, Guo P. Advances in Neural Networks-ISNN 2004, PT1, 3173: 616.
[5] Li X R, Lin Y T, Qiu K B. Research in Astronomy and Astrophysics, 2019, 19(8): 111.
[6] Zhao Z, Wei J Y, Jiang B. Advances in Astronomy, 2022, 2022: 4489359.
[7] 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): 16.
[8] Zou Z Q, Zhu T C, Xu L Z, et al. Publications of the Astronomical Society of the Pacific, 2020, 132(1010): 044503.
[9] He K, Zhang X, Ren S, et al. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016: 770.
[10] Cui X Q, Zhao Y H, Chu Y Q, et al. Research in Astronomy and Astrophysics, 2012, 12(9): 1197.
[11] Wang Q, Wu B, Zhu P, et al. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2020: 11531.
[12] Cui B, Ye Z, Bai Z. Chinese Astronomy and Astrophysics, 2009, 33(1): 99.
[13] Liu W, Zhu M, Dai C, et al. Monthly Notices of the Royal Astronomical Society, 2019, 483(4): 4774.
|
| [1] |
YU Shui1, HUAN Ke-wei1*, LIU Xiao-xi2, WANG Lei1. Quantitative Analysis Modeling of Near Infrared Spectroscopy With
Parallel Convolution Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(06): 1627-1635. |
| [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] |
WANG Zi-xuan1, YANG Liang2, 3, 4*, HUANG Ling-xia2, HE Yong4, ZHAO Li-hua3, ZHAN Peng-fei3. Nondestructive Determination of TSS Content in Postharvest Mulberry Fruits Using Hyperspectral Imaging and Deep Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(06): 1724-1730. |
| [4] |
LIU Zong-yi1, 2, ZHANG Cai-hong1, 2, JIANG Jian-kang1, 2, SHEN Bin-guo3, DING Yan-fei1, 2, ZHANG Lei-lei1, 2*, ZHU Cheng1, 2*. Rapid Determination of Total Flavonoids in Dendrobium Officinale Based on Raman Spectroscopy and CNN-LSTM Deep Learning Method[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(04): 1018-1024. |
| [5] |
YANG Wan-qi1, 2, LI Zhi-qi1, 3, LI Fu-sheng1, 2*, LÜ Shu-bin1, 2, FAN Jia-jing1, 2. A Combined CARS and 1D-CNN Method for the Analysis of Heavy Metals Exceedances in Soil by XRF Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(03): 670-674. |
| [6] |
TANG Jie1, LUO Yan-bo2, LI Xiang-yu2, CHEN Yun-can1, WANG Peng1, LU Tian3, JI Xiao-bo4, PANG Yong-qiang2*, ZHU Li-jun1*. Study on One-Dimensional Convolutional Neural Network Model Based on Near-Infrared Spectroscopy Data[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(03): 731-736. |
| [7] |
LI Guo-hou1, LI Ze-xu1, JIN Song-lin1, ZHAO Wen-yi2, PAN Xi-peng3, LIANG Zheng4, QIN Li5, ZHANG Wei-dong1*. Mix Convolutional Neural Networks for Hyperspectral Wheat Variety
Discrimination[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(03): 807-813. |
| [8] |
SUN He-yang1, ZHOU Yue1, 2, LI Si-jia1, 2, LI Li1, YAN Ling-tong1, FENG Xiang-qian1*. Identification of Ancient Ceramic by Convolution Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(02): 354-358. |
| [9] |
ZHANG Hai-liang1, ZHOU Xiao-wen1, LIU Xue-mei2*, LUO Wei2, ZHAN Bai-shao2, PAN Fan3. Freshness Identification of Turbot Based on Convolutional Neural
Network and Hyperspectral Imaging Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(02): 367-371. |
| [10] |
MENG Xing-zhi, LIU Ya-qiu*, LIU Li-na. Raman Spectroscopy Combined With the WGANGP-ResNet Algorithm to Identify Pathogenic Species[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(02): 542-547. |
| [11] |
ZHANG Fan1, WANG Wen-xiu1, WANG Chun-shan2, ZHOU Ji2, PAN Yang3, SUN Jian-feng1*. Study on Hyperspectral Detection of Potato Dry Rot in Gley Stage Based on Convolutional Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(02): 480-489. |
| [12] |
LI Xin-ting, ZHANG Feng, FENG Jie*. Convolutional Neural Network Combined With Improved Spectral
Processing Method for Potato Disease Detection[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 215-224. |
| [13] |
LAN Yan1，WANG Wu1，XU Wen2，CHAI Qin-qin1*，LI Yu-rong1，ZHANG Xun2. Discrimination of Planting and Tissue-Cultured Anoectochilus Roxburghii Based on SMOTE and Inception-CNN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 158-163. |
| [14] |
WANG Qi-biao1, HE Yu-kai1, LUO Yu-shi1, WANG Shu-jun1, XIE Bo2, DENG Chao2*, LIU Yong3, TUO Xian-guo3. Study on Analysis Method of Distiller's Grains Acidity Based on
Convolutional Neural Network and Near Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(12): 3726-3731. |
| [15] |
HUANG Meng-qiang1, KUANG Wen-jian2, 3*, LIU Xiang1, HE Liang4. Quantitative Analysis of Cotton/Polyester/Wool Blended Fiber Content by Near-Infrared Spectroscopy Based on 1D-CNN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(11): 3565-3570. |
|
|
|
|
