|
|
|
|
|
|
Extracting Convolutional Features of WDMS Spectra with Anti Bayesian Learning Paradigm |
WANG Wen-yu, GUO Ge-lin, MA Chun-yu, JIANG Bin* |
School of Mechanical, Electrical & Information Engineering, Shandong University, Weihai, Weihai 264209, China |
|
|
Abstract In the task of White Dwarf+Main Sequence (WDMS) finding in massive spectral data release, convolution can significantly improve the classification accuracy by extracting hierarchical, translational-invariant features. In this paper, by designing one dimensional convolutional neural network (1-D CNN) which was further trained in a discriminative, supervised way, 12 kernels with stable numerical distributions were produced, helping to generate spectral feature maps of WDMS. To solve the problem brought by biased sampling in the WDMS training set, we proposed a learning principle called Anti-Bayesian Learning Paradigm (ALP) which was built on the basis of order statistics by implying a comparatively looser prior distribution of spectral types. And in the way of separating training spectra into several groups according to their signal-to-noise ratios (SNR), we analyzed the robustness of convolutional extraction process to spectral noise. Experimental results indicated that, (1) WDMS classification with 1-D CNN and ALP reached the accuracy of 99.0%±0.3%, which outperformed the classic PCA+SVM model. (2) Pooling after convolution operations relieved the negative impact of spectral noise by lowering resolution. (3) When the SNR was less than 3, more epochs were required to learn stable kernels; when the SNR was between 3 and 6, the spectral convolutional features was stable; when the SNR was greater than 6, the convolution process acquired higher stability to eliminate the negative impact of SNR on model performance.
|
Received: 2017-02-08
Accepted: 2017-10-16
|
|
Corresponding Authors:
JIANG Bin
E-mail: emilyMCY@163.com
|
|
[1] REN Juan-juan, LUO A-li, ZHAO Yong-heng(任娟娟,罗阿理,赵永恒). Progress in Astronomy(天文学进展), 2014, 32(4): 462.
[2] Yu H, Xie T, Paszczynski S, et al. IEEE Transactions on Industrial Electronics, 2011, 58(12): 5438.
[3] Yue J, Zhao W, Mao S, et al. Remote Sensing Letters, 2015, 6(6): 468.
[4] Sun M, Song Z, Jiang X, et al. Neurocomputing, 2017, 224: 96.
[5] Bashkirova D. BioNanoScience, 2016, 6(3): 246.
[6] Guohao L Y U, Hui Y I N, Xinyan Y U, et al. IEICE TRANSACTIONS on Information and Systems, 2016, 99(8): 2190.
[7] Krizhevsky Alex, Sutskever I, Hinton G E. Advances in Neural Information Processing Systems, 2012, 25(2): 2012.
[8] Oommen, John B, Thomas A. Pattern Recognition, 2014, 47(1): 40.
[9] Du Wei, Luo A-li, Zhao Yong-heng. Spectroscopy and Spectral Analysis, 2012, 32(7): 1886.
[10] Rebassa-Mansergas A, Nebot Gomez-Moran A, Schreiber M R, et al. Monthly Notices of the Royal Astronomical Society, 2012, 419(1):806. |
[1] |
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. |
[2] |
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. |
[3] |
CAI Jian-rong1, 2, HUANG Chu-jun1, MA Li-xin1, ZHAI Li-xiang1, GUO Zhi-ming1, 3*. Hand-Held Visible/Near Infrared Nondestructive Detection System for Soluble Solid Content in Mandarin by 1D-CNN Model[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(09): 2792-2798. |
[4] |
PU Shan-shan, ZHENG En-rang*, CHEN Bei. Research on A Classification Algorithm of Near-Infrared Spectroscopy Based on 1D-CNN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(08): 2446-2451. |
[5] |
CHEN Dong-ying1, 2, ZHANG Hao1, 2*, ZHANG Zi-long1, YU Mu-xin1, CHEN Lu3. Research on the Origin Traceability of Honeysuckle Based on Improved 1D-VD-CNN and Near-Infrared Spectral Data[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(05): 1471-1477. |
[6] |
PAN Ke-yu1, 2, ZHU Ming-yao1, 2, WANG Yi-meng1, 2, XU Yang1, CHI Ming-bo1, 2*, WU Yi-hui1, 2*. Research on the Influence of Modulation Depth of Phase Sensitive
Detection on Stimulated Raman Signal Intensity and
Signal-to-Noise Ratio[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(04): 1068-1074. |
[7] |
CHEN Qing1, TANG Bin1, 2*, LONG Zou-rong1, 2, MIAO Jun-feng1, HUANG Zi-heng1, DAI Ruo-chen1, SHI Sheng-hui1, ZHAO Ming-fu1, ZHONG Nian-bing1. Water Quality Classification Using Convolution Neural Network Based on UV-Vis Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(03): 731-736. |
[8] |
LIU Ye-kun, HAO Xiao-jian*, YANG Yan-wei, HAO Wen-yuan, SUN Peng, PAN Bao-wu. Quantitative Analysis of Soil Heavy Metal Elements Based on Cavity
Confinement LIBS Combined With Machine Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(08): 2387-2391. |
[9] |
WANG Yue1, 3, 4, CHEN Nan1, 2, 3, 4, WANG Bo-yu1, 5, LIU Tao1, 3, 4*, XIA Yang1, 2, 3, 4*. Fourier Transform Near-Infrared Spectral System Based on Laser-Driven Plasma Light Source[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1666-1673. |
[10] |
ZHAO Yong1, HE Men-yuan1, WANG Bo-lin2, ZHAO Rong2, MENG Zong1*. Classification of Mycoplasma Pneumoniae Strains Based on
One-Dimensional Convolutional Neural Network and
Raman Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(05): 1439-1444. |
[11] |
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. |
[12] |
TAN Ai-ling1, CHU Zhen-yuan1, WANG Xiao-si1, ZHAO Yong2*. Detection of Pearl Powder Adulteration Based on Raman Spectroscopy and DCGAN Data Enhancement[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 769-775. |
[13] |
HU Li-hong1, ZHANG Jin-tong1, WANG Li-yun2, ZHOU Gang3, WANG Jiang-yong1*, XU Cong-kang1*. Optimization of Working Parameters of Glow Discharge Optical Emission Spectrometry of High Barrier Aluminum Plastic Film[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 954-960. |
[14] |
CUI Fang-xiao1, ZHAO Yue2, MA Feng-xiang2, WU Jun1*, WANG An-jing1, LI Da-cheng1, LI Yang-yu1. Optimization of FTIR Passive Remote Sensing Signal-to-Noise Ratio and Its Application in SF6 Leak Detection in Transform Substation[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(05): 1436-1440. |
[15] |
WANG Jing-jing1, 2, TAN Tu1*, WANG Gui-shi1, ZHU Gong-dong1, XUE Zheng-yue1, 2, LI Jun1, 2, LIU Xiao-hai1, 2, GAO Xiao-ming1, 2. Research on All-Fiber Dual-Channel Atmospheric Greenhouse Gases Laser Heterodyne Detection Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(02): 354-359. |
|
|
|
|