Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 57, Issue 14, 141501(2020)

Facial Expression Classification Based on Ensemble Convolutional Neural Network

Tao Zhou1, Xiaoqi Lü1,2,3、*, Guoyin Ren1, Yu Gu1,3, Ming Zhang1,4, and Jing Li1
Author Affiliations
  • 1Inner Mongolia Key Laboratory of Pattern Recognition and Intelligent Image Progressing, School of Information Engineering, Inner Mongolia University of Science and Technology, Baotou, Inner Mongolia 0 14010, China
  • 2School of Information Engineering, Inner Mongolia University of Technology, Hohhot, Inner Mongolia 0 10051, China
  • 3School of Computer Engineering and Science, Shanghai University, Shanghai 200444, China
  • 4Information Science and Technology College, Dalian Maritime University, Dalian, Liaoning 116026, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (16)
    Equations (0)
    References (25)
    Cited By (4)
    Tools
    Paper Information
    Topics
    Figures & Tables(16)
    Model structure of VGGNet-19GP

    Fig. 1. Model structure of VGGNet-19GP

    Download full size
    Model structure of EnsembleNet

    Fig. 2. Model structure of EnsembleNet

    Download full size
    Example images from the CK+ dataset

    Fig. 3. Example images from the CK+ dataset

    Download full size
    Example images from the FER2013 dataset

    Fig. 4. Example images from the FER2013 dataset

    Download full size
    Schematic of the training set data enhancement

    Fig. 5. Schematic of the training set data enhancement

    Download full size
    Schematic of the test set data enhancement

    Fig. 6. Schematic of the test set data enhancement

    Download full size
    Result graphs of simple average experiment. Comparison of the average accuracy of EnsembleNet, ResNet-18, and VGGNet-19GP under (a) PublicTest and (b) PrivateTest with the increase of epoch

    Fig. 7. Result graphs of simple average experiment. Comparison of the average accuracy of EnsembleNet, ResNet-18, and VGGNet-19GP under (a) PublicTest and (b) PrivateTest with the increase of epoch

    Download full size
    Result graphs of weighted average.Fluctuations of EnsembleNet with the change of ResNet-18 weights under (a) PublicTest and (b) PrivateTest, and the comparison with ResNet-18 and VGGNet-19GP

    Fig. 8. Result graphs of weighted average.Fluctuations of EnsembleNet with the change of ResNet-18 weights under (a) PublicTest and (b) PrivateTest, and the comparison with ResNet-18 and VGGNet-19GP

    Download full size
    VGGNet-19GP, ResNet-18, and EnsembleNet accuracy curves on PublicTest dataset

    Fig. 9. VGGNet-19GP, ResNet-18, and EnsembleNet accuracy curves on PublicTest dataset

    Download full size
    VGGNet-19GP, ResNet-18, and EnsembleNet accuracy curves on PrivateTest dataset

    Fig. 10. VGGNet-19GP, ResNet-18, and EnsembleNet accuracy curves on PrivateTest dataset

    Download full size
    VGGNet-19GP, ResNet-18, and EnsembleNet accuracy curves on CK+ dataset

    Fig. 11. VGGNet-19GP, ResNet-18, and EnsembleNet accuracy curves on CK+ dataset

    Download full size
    Confusion matrix of EnsembleNet on the FER2013 dataset. (a) PublicTest Confusion Matrix; (b) PrivateTest Confusion Matrix

    Fig. 12. Confusion matrix of EnsembleNet on the FER2013 dataset. (a) PublicTest Confusion Matrix; (b) PrivateTest Confusion Matrix

    Download full size
    Confusion matrix of EnsembleNet on the CK+ dataset

    Fig. 13. Confusion matrix of EnsembleNet on the CK+ dataset

    Download full size
    Examples of correct classification and misclassification of PrivateTest

    Fig. 14. Examples of correct classification and misclassification of PrivateTest

    Download full size

    • Table 1. Average accuracy on the FER2013 and CK+ datasets%

      View table

      Table 1. Average accuracy on the FER2013 and CK+ datasets%

      ModelFER2013CK+
      Public_Avg_AccPrivate_Avg_AccAvg_Acc
      VGGNet-19GP70.61671.84891.107
      ResNet-1871.32772.27192.845
      EnsembleNet71.69773.85497.611

    • Table 2. Comparison of proposed model with existing methods on the FER2013 and CK+datasets

      View table

      Table 2. Comparison of proposed model with existing methods on the FER2013 and CK+datasets

      SourceMethodDataaugmentedDropoutAccuracy /%
      FER2013CK+
      Ref. [8]Pre-processing+5_Layer_CNN---97.75
      Ref. [9]Landmark+5_Layer_CNN---97.25
      Ref. [10]CSACNN√--97.45
      Ref. [11]7_CNN---81.50
      Ref. [11]7_CNN-√82.90
      Ref. [11]7_CNN√84.42
      Ref. [11]7_CNN√√84.55
      Ref. [12]Cross-connect LeNet-5---83.74
      Ref. [13]Siamese network with multiple channels---92.06
      Ref. [14]Multi-resolution feature fusion---92.10
      Ref. [15]Local feature fusion---94.56
      Ref. [16]Parallel CNN√-65.694.03
      Ref. [17]Ensemble CNNs+L2_Loss--71.16-
      Ref. [23]CNN+FACS+AU--72.198.62
      Ref. [24]11_Layer_CNN-√65.3-
      Ref. [25]Fully-convolution neural network--66-
      Ref. [26]8_CNN(filters decreases with net depth)--65-
      ProposedVGGNet-19GP√√71.84891.107
      ProposedResNet-18√√72.27192.845
      ProposedEnsembleNet√√73.85497.611
    Tools

    Get Citation

    Copy Citation Text

    Tao Zhou, Xiaoqi Lü, Guoyin Ren, Yu Gu, Ming Zhang, Jing Li. Facial Expression Classification Based on Ensemble Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2020, 57(14): 141501

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Set citation alerts for article
    Save article for my favorites
    Paper Information

    Category: Machine Vision

    Received: Oct. 8, 2019

    Accepted: Nov. 26, 2019

    Published Online: Jul. 28, 2020

    The Author Email: Xiaoqi Lü (lxiaoqi@imut.edu.cn)

    DOI:10.3788/LOP57.141501

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology