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. 58, Issue 6, 610019(2021)

Chinese Food Recognition Model Based on Improved Residual Network

Deng Zhiliang and Li Lei*
Author Affiliations
  • School of Automation, Nanjing University of Information Science & Technology, Nanjing, Jiangsu 210044, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (15)
    Equations (0)
    References (18)
    Cited By (2)
    Tools
    Paper Information
    Topics
    Figures & Tables(15)
    Model structure

    Fig. 1. Model structure

    Download full size
    RNA network structure

    Fig. 2. RNA network structure

    Download full size
    Convolution structure of RNA-A

    Fig. 3. Convolution structure of RNA-A

    Download full size
    Convolution structure of RNA-B

    Fig. 4. Convolution structure of RNA-B

    Download full size
    Convolution structure of RNA-C

    Fig. 5. Convolution structure of RNA-C

    Download full size
    Basic structure of attention mechanism module

    Fig. 6. Basic structure of attention mechanism module

    Download full size
    Schematic of TL learning

    Fig. 7. Schematic of TL learning

    Download full size
    Samples in Food208 dataset

    Fig. 8. Samples in Food208 dataset

    Download full size
    Samples in Food292 dataset

    Fig. 9. Samples in Food292 dataset

    Download full size
    Time required for training different CNNs on Food208 and Food292 datasets

    Fig. 10. Time required for training different CNNs on Food208 and Food292 datasets

    Download full size

    • Table 1. Image numbers of training dataset and test dataset for different datasets

      View table

      Table 1. Image numbers of training dataset and test dataset for different datasets

      DatasetTrainingTest
      Food20812801620214
      Food2927008217521

    • Table 2. Model parameter setting

      View table

      Table 2. Model parameter setting

      ParameterContent
      Input size224 pixel×224 pixel
      Epoch200
      Batch size90
      OptimizerAdam
      Dropout parameter0.4

    • Table 3. Recognition accuracies of different CNNs on Food208 and Food292 datasets

      View table

      Table 3. Recognition accuracies of different CNNs on Food208 and Food292 datasets

      Convolutional neural networkFood208Food292
      Inception-v3[16]70.52%80.65%
      ResNet-18[17]74.64%82.41%
      Inception-v4[18]79.51%83.16%
      Inception-ResNet-v1[18]79.93%85.53%
      Inception-ResNet-v2[18]80.36%86.10%
      RNA83.66%90.31%

    • Table 4. Accuracies of different CNNs when loss function is formula (2)

      View table

      Table 4. Accuracies of different CNNs when loss function is formula (2)

      Convolutional neural networkFood208Food292
      Inception-v369.52%79.53%
      ResNet-1873.15%81.12%
      Inception-v478.30%81.91%
      Inception-ResNet-v178.21%84.46%
      Inception-ResNet-v278.95%84.95%
      RNA82.50%89.10%

    • Table 5. Accuracies obtained by RNA-TL under different epochs

      View table

      Table 5. Accuracies obtained by RNA-TL under different epochs

      EpochFood208Food292
      5078.95%85.63%
      10080.22%87.02%
      15082.13%89.23%
      20083.60%90.31%
      25083.62%90.32%
    Tools

    Get Citation

    Copy Citation Text

    Deng Zhiliang, Li Lei. Chinese Food Recognition Model Based on Improved Residual Network[J]. Laser & Optoelectronics Progress, 2021, 58(6): 610019

    Download Citation

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

    Category: Image Processing

    Received: Jul. 20, 2020

    Accepted: --

    Published Online: Mar. 11, 2021

    The Author Email: Lei Li (466743943@qq.com)

    DOI:10.3788/LOP202158.0610019

    Topics

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