• Acta Photonica Sinica
  • Vol. 50, Issue 10, 1010001 (2021)
Feng XIONG1, Di HE1, Yujie LIU1, Meijie QI1, Peng GAO1, Zhoufeng ZHANG2、*, and Lixin LIU1、2、*
Author Affiliations
  • 1School of Physics and Optoelectronic Engineering,Xidian University,Xi'an 710071,China
  • 2CAS Key Laboratory of Spectral Imaging Technology,Xi'an Institute of Optics and Precision Mechanics,Chinese Academy of Sciences,Xi'an 710119,China
  • show less
    DOI: 10.3788/gzxb20215010.1010001 Cite this Article
    Feng XIONG, Di HE, Yujie LIU, Meijie QI, Peng GAO, Zhoufeng ZHANG, Lixin LIU. Classification of Pneumonia Images Based on Improved VGG19 Convolutional Neural Network(Invited)[J]. Acta Photonica Sinica, 2021, 50(10): 1010001 Copy Citation Text show less
    Examples of lung X-ray images
    Fig. 1. Examples of lung X-ray images
    Schematic diagram of image preprocessing
    Fig. 2. Schematic diagram of image preprocessing
    Basic principle of bilinear interpolation
    Fig. 3. Basic principle of bilinear interpolation
    The size of the original X-ray image is adjusted to 224×224
    Fig. 4. The size of the original X-ray image is adjusted to 224×224
    Lung X-ray image before and after linear transformation
    Fig. 5. Lung X-ray image before and after linear transformation
    Image contrast enhancement comparison before and after CLAHE processing
    Fig. 6. Image contrast enhancement comparison before and after CLAHE processing
    Diagram of improved VGG19 model
    Fig. 7. Diagram of improved VGG19 model
    Workflow of the improved VGG19 framework
    Fig. 8. Workflow of the improved VGG19 framework
    Model accuracy changing with the number of test
    Fig. 9. Model accuracy changing with the number of test
    The performance evaluation of three best models with highest accuracy
    Fig. 10. The performance evaluation of three best models with highest accuracy
    ModelLowest accuracyHighest accuracyAverage accuracyAccuracy variance
    VGG1984.1%89.0%85.9%1.73
    SVM(linear)-based VGG1985.2%86.6%85.9%0.31
    XGBoost-based VGG1987.2%89.4%88.2%0.37
    Table 1. Performance comparison between VGG19 and two improved VGG19 models
    Feng XIONG, Di HE, Yujie LIU, Meijie QI, Peng GAO, Zhoufeng ZHANG, Lixin LIU. Classification of Pneumonia Images Based on Improved VGG19 Convolutional Neural Network(Invited)[J]. Acta Photonica Sinica, 2021, 50(10): 1010001
    Download Citation