• Spectroscopy and Spectral Analysis
  • Vol. 42, Issue 6, 1881 (2022)
Ya-kun LU1、*, Bo QIU1、1; *;, A-li LUO2、2;, Xiao-yu GUO1、1;, Lin-qian WANG1、1;, Guan-long CAO1、1;, Zhong-rui BAI2、2;, and Jian-jun CHEN2、2;
Author Affiliations
  • 11. Hebei University of Technology, Tianjin 300400, China
  • 22. National Astronomical Observatory, Chinese Academy of Sciences, Beijing 100012, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2022)06-1881-05 Cite this Article
    Ya-kun LU, Bo QIU, A-li LUO, Xiao-yu GUO, Lin-qian WANG, Guan-long CAO, Zhong-rui BAI, Jian-jun CHEN. Classification of 2D Stellar Spectra Based on FFCNN[J]. Spectroscopy and Spectral Analysis, 2022, 42(6): 1881 Copy Citation Text show less
    Spectral image of LAMOST object
    Fig. 1. Spectral image of LAMOST object
    An image of 2D spectrum
    Fig. 2. An image of 2D spectrum
    The structure of FFCNN
    Fig. 3. The structure of FFCNN
    The influence of convolution kernel with different length on the accuracy and the time spent in the model training
    Fig. 4. The influence of convolution kernel with different length on the accuracy and the time spent in the model training
    Classification results of all samples in the testing set
    Fig. 5. Classification results of all samples in the testing set
    The comparison of classification results of using 2D and 1D spectra
    Fig. 6. The comparison of classification results of using 2D and 1D spectra
    FGK
    Precision0.8760.7920.885
    Recall0.7630.9030.742
    F1-score0.8160.8430.807
    Accuracy0.829
    Table 1. The precision, recall, F1-score and the accuracy of FFCNN model
    Ya-kun LU, Bo QIU, A-li LUO, Xiao-yu GUO, Lin-qian WANG, Guan-long CAO, Zhong-rui BAI, Jian-jun CHEN. Classification of 2D Stellar Spectra Based on FFCNN[J]. Spectroscopy and Spectral Analysis, 2022, 42(6): 1881
    Download Citation