• Spectroscopy and Spectral Analysis
  • Vol. 42, Issue 5, 1439 (2022)
Yong ZHAO1、1;, Men-yuan HE1、1;, Bo-lin WANG2、2;, Rong ZHAO2、2;, and Zong MENG1、1; *;
Author Affiliations
  • 11. School of Electrical Engineering, Yanshan University, The Key Laboratory of Measurement Technology and Instruments of Hebei Province, Qinhuangdao 066004, China
  • 22. School of Information Science and Engineering, Yanshan University, The Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, Qinhuangdao 066004, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2022)05-1439-06 Cite this Article
    Yong ZHAO, Men-yuan HE, Bo-lin WANG, Rong ZHAO, Zong MENG. Classification of Mycoplasma Pneumoniae Strains Based on One-Dimensional Convolutional Neural Network and Raman Spectroscopy[J]. Spectroscopy and Spectral Analysis, 2022, 42(5): 1439 Copy Citation Text show less
    Raman spectra of two strains of mycoplasma pneumoniae
    Fig. 1. Raman spectra of two strains of mycoplasma pneumoniae
    Raman spectral data enhancement
    Fig. 2. Raman spectral data enhancement
    Original spectrum and spectrum with noise(a): Original spectrum; (b): Spectrum with Gaussian noise; (c): Spectrum with poisson noise; (d): Spectrum with Multiplicative noise
    Fig. 3. Original spectrum and spectrum with noise
    (a): Original spectrum; (b): Spectrum with Gaussian noise; (c): Spectrum with poisson noise; (d): Spectrum with Multiplicative noise
    1-D convolutional neural network Raman spectra classification model
    Fig. 4. 1-D convolutional neural network Raman spectra classification model
    Accuracy and loss curve of 1D-CNN model
    Fig. 5. Accuracy and loss curve of 1D-CNN model
    Confusion matrix of different algorithms with three kinds of noises(a): Gaussian noise; (b): Poisson noise; (c): Multiplicative noise
    Fig. 6. Confusion matrix of different algorithms with three kinds of noises
    (a): Gaussian noise; (b): Poisson noise; (c): Multiplicative noise
    ROC curves of different algorithms with three kinds of noises(a): Gaussian noise; (b): Poisson noise; (c): Multiplicative noise
    Fig. 7. ROC curves of different algorithms with three kinds of noises
    (a): Gaussian noise; (b): Poisson noise; (c): Multiplicative noise
    Classification results with different algorithms for different intensity of noise
    Fig. 8. Classification results with different algorithms for different intensity of noise
    高斯噪声泊松噪声乘性噪声
    正确率灵敏度特异性正确率灵敏度特异性正确率灵敏度特异性
    LDA62.0067.8654.5560.0060.0752.2764.0069.6456.82
    KNN62.0058.9365.9168.0062.5075.0066.0051.7984.09
    SVM65.0058.9372.7369.0062.5077.2767.0060.7175.00
    CNN98.0096.43100.0097.0098.2195.4597.00100.0093.18
    Table 1. Comparison of classification results with different algorithms
    Yong ZHAO, Men-yuan HE, Bo-lin WANG, Rong ZHAO, Zong MENG. Classification of Mycoplasma Pneumoniae Strains Based on One-Dimensional Convolutional Neural Network and Raman Spectroscopy[J]. Spectroscopy and Spectral Analysis, 2022, 42(5): 1439
    Download Citation