Laser & Optoelectronics Progress, Volume. 55, Issue 3, 031103(2018)

Adaptive Window Local Matching Algorithm Based on HSV Color Space

Xiu Su, Xiaodong Chen*, Huaiyuan Xu, Haitao Liang, Yilin Liu, Yi Wang, Weifeng Li, and Daoyin Yu
Author Affiliations
  • Key Laboratory of the Ministry of Education on Optoelectronic Information Technology, School of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China
  • show less
    Figures & Tables(9)
    Block diagram of proposed algorithm
    Color component grayscale of Tsukuba. (a) R component; (b) G component; (c) B component; (d) H component; (e) S component; (f) V component
    Comparison of different aggregation regions. (a) Fixed region; (b) adaptive aggregation region; (c) aggregation region selected by proposed method
    (a) Reference images; (b) actual disparity maps; simulation results obtained by (c) proposed algorithm, (d) adaptive stereo matching method and (e) ASSW method for Middlebury stereo pairs
    PSNR comparison of proposed algorithm with other similar methods
    Running time comparison of proposed algorithm and other similar methods
    Robustness tests for parameters (a) Lmax and (b) τ
    • Table 1. Average percentage of mismatched pixels with threshold of 1%

      View table

      Table 1. Average percentage of mismatched pixels with threshold of 1%

      AlgorithmTsukubaVenusTeddyConesAverage error
      NooccAllDiscNooccAllDiscNooccAllDiscNooccAllDisc
      Proposed1.581.865.830.270.462.406.1011.3014.502.718.427.505.24
      ASSW[13]1.812.177.850.320.513.737.0212.5017.403.218.408.996.16
      Adapt weight[12]1.381.856.900.711.196.137.8813.3018.603.979.798.266.67
      Cost filter[18]1.511.857.610.200.392.426.1611.8016.002.718.247.665.55
      Variable cross[19]1.992.656.770.620.963.209.7515.1018.206.2812.7012.907.60
      SMPF[20]0.981.535.310.250.692.609.9314.5022.606.5113.1014.807.73
      GC+occ[21]1.192.016.241.642.196.7511.2017.4019.805.3612.4013.008.26
      VSW (virtualsupport window)[22]1.621.886.980.470.813.408.6713.3018.003.378.828.126.29
    • Table 2. Average percentage of mismatched pixels with threshold of 0.5%

      View table

      Table 2. Average percentage of mismatched pixels with threshold of 0.5%

      AlgorithmTsukubaVenusTeddyConesAverage error
      NooccAllDiscNooccAllDiscNooccAllDiscNooccAllDisc
      Proposed13.7014.0016.007.087.4911.5012.3018.5024.008.9114.9015.1013.60
      ASSW[13]11.9012.5017.006.667.1112.3012.7019.1026.609.0114.5016.3013.80
      Adapt weight[12]18.1018.8018.607.778.4015.8017.6023.9034.0014.0019.7020.6018.10
      Variable cross[19]24.5025.1021.509.039.5913.8018.8025.1031.4016.1022.1022.4019.90
      SMPF[20]19.5020.1016.608.909.5615.1020.5025.9038.2013.9020.5026.2019.60
      GC+occ[21]6.107.1114.6010.7011.3016.9023.7030.1034.6012.2019.2021.9017.40
      VSW[22]19.2019.5018.508.178.6513.2017.4023.2031.4013.1018.3020.4017.60
    Tools

    Get Citation

    Copy Citation Text

    Xiu Su, Xiaodong Chen, Huaiyuan Xu, Haitao Liang, Yilin Liu, Yi Wang, Weifeng Li, Daoyin Yu. Adaptive Window Local Matching Algorithm Based on HSV Color Space[J]. Laser & Optoelectronics Progress, 2018, 55(3): 031103

    Download Citation

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

    Category: Imaging Systems

    Received: Sep. 13, 2017

    Accepted: --

    Published Online: Sep. 10, 2018

    The Author Email: Chen Xiaodong (xdchen@tju.edu.cn)

    DOI:10.3788/LOP55.031103

    Topics