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. 56, Issue 8, 081007(2019)

Methods for Location and Recognition of Chess Pieces Based on Convolutional Neural Network

Xie Han*, Rong Zhao**, and Fusheng Sun***
Author Affiliations
  • Department of Data Science and Technology, North University of China, Taiyuan, Shanxi 030051, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (15)
    Equations (0)
    References (19)
    Cited By (2)
    Tools
    Paper Information
    Topics
    Figures & Tables(15)
    Flow chart of whole algorithm

    Fig. 1. Flow chart of whole algorithm

    Download full size
    Flow chart of algorithm for location of chess pieces

    Fig. 2. Flow chart of algorithm for location of chess pieces

    Download full size
    Chessboard pretreatment. (a) Perspective transformation picture; (b) ROI picture

    Fig. 3. Chessboard pretreatment. (a) Perspective transformation picture; (b) ROI picture

    Download full size
    Linear mixture picture

    Fig. 4. Linear mixture picture

    Download full size
    Location picture of chess pieces

    Fig. 5. Location picture of chess pieces

    Download full size
    Flow chart of recognition algorithm

    Fig. 6. Flow chart of recognition algorithm

    Download full size
    Network structure

    Fig. 7. Network structure

    Download full size
    Examples of chess data

    Fig. 8. Examples of chess data

    Download full size
    Training and verification results of proposed method. (a) Training accuracy and validation accuracy; (b) training loss and verification loss

    Fig. 9. Training and verification results of proposed method. (a) Training accuracy and validation accuracy; (b) training loss and verification loss

    Download full size
    Recognition results of chess pieces based on CNN. (a) Partial experimental results 1; (b) partial experimental results 2

    Fig. 10. Recognition results of chess pieces based on CNN. (a) Partial experimental results 1; (b) partial experimental results 2

    Download full size
    Comparison of experimental results

    Fig. 11. Comparison of experimental results

    Download full size

    • Table 1. Configuration information and data of network structure (Conv1-Conv4 layout data)

      View table

      Table 1. Configuration information and data of network structure (Conv1-Conv4 layout data)

      ConvConv1Conv2Conv3Conv4
      Data_Size100×100×348×48×3224×24×6424×24×128
      Conv: Num_Filter3264128128
      Conv: padding0211
      Conv: Filter_Size5×5×35×5×323×3×6424×24×128
      Conv: stride1111
      Data_Size after convolution96×96×3248×48×6424×24×12824×24×128
      ActivationReLUReLUReLUReLU
      Data_Size after activation96×96×3248×48×6424×24×12824×24×128
      Pooling: Kernel_Size2×22×22×2
      Pooling: stride222
      Data_Size after pooling48×48×3224×24×6424×24×12812×12×128
      LRN(Data_Size)48×48×3224×24×6424×24×12812×12×128

    • Table 2. Configuration information and data of network structure(FC1-FC3 layout data)

      View table

      Table 2. Configuration information and data of network structure(FC1-FC3 layout data)

      FCFC1FC2FC3
      Data12×12×1281024512
      Data after FC1024102414
      ActivationReLUReLU
      Data after activation10241024
      Dropout Kept_prob0.50.5
      Data after dropout fitting1024512

    • Table 3. Location experiment of chess pieces

      View table

      Table 3. Location experiment of chess pieces

      PieceTime ofsegmentation /sCoordinate of imageCalculated coordinateActual coordinateError /mm
      Col /pixelRow /pixelX /mmY /mmX' /mmY' /mm
      Red_Car0.1731292.88972.66-56.816-101.251-56.7-100.21.09
      Red_House0.1771286.983921.603-57.532-32.759-57.5-61.61.15
      Red_Ele0.1911279.039869.282-57.568-21.625-57.4-21.50.2
      Red_Knight0.2071273.411824.562-57.6515.628-57.515.60.15
      Marshal0.1821266.772777.076-57.10756.284-57.156.40.12
      Red_Gun0.1861189.585919.27320.505-62.27920.5-61.70.58
      Red_Pawn0.181128.461776.13261.96455.66761.755.80.29
      Green_Pawn0.185986.305614.135185.19210.372186.2210.81.1
      Green_Gun0.182943.496979.248219.448-64.996219.6-650.15
      General0.179854.364771.18299.64657.123299.557.50.4
      Green_Knight0.184850.955814.87300.59719.501300.119.20.36
      Green_Ele0.183849.73863.732298.672-21.476299.1-21.60.45
      Green_House0.183845.265918.192298.95-65.306298.8-65.90.61
      Green_Car0.186843.988956.32298.05-101.029297.6-101.30.53
      Total piece0.208------0.51

    • Table 4. Comparison of experimental results

      View table

      Table 4. Comparison of experimental results

      Experimental dataProposed methodRef. [3]Ref. [14]
      Location time /s0.2120.484-
      Location error /mm0.51-0.87
    Tools

    Get Citation

    Copy Citation Text

    Xie Han, Rong Zhao, Fusheng Sun. Methods for Location and Recognition of Chess Pieces Based on Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2019, 56(8): 081007

    Download Citation

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

    Category: Image Processing

    Received: Oct. 17, 2018

    Accepted: Nov. 22, 2018

    Published Online: Jul. 26, 2019

    The Author Email: Han Xie (290949559@qq.com), Zhao Rong (tm_zhaorong@126.com), Sun Fusheng (sfs2699@163.com)

    DOI:10.3788/LOP56.081007

    Topics

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