• Laser & Optoelectronics Progress
  • Vol. 59, Issue 8, 0810011 (2022)
Wenjie Yu1, Song Ye2, Yu Guo1、*, and Jian Guo1
Author Affiliations
  • 1School of Automation, Nanjing University of Science & Technology, Nanjing , Jiangsu 210094, China
  • 2The Third Construction Co., Ltd. of China Construction Eighth Engineering Division, Nanjing , Jiangsu 210023, China
  • show less
    DOI: 10.3788/LOP202259.0810011 Cite this Article Set citation alerts
    Wenjie Yu, Song Ye, Yu Guo, Jian Guo. Stereo Matching Algorithm Based on Improved Census Transform and Multi-Feature Fusion[J]. Laser & Optoelectronics Progress, 2022, 59(8): 0810011 Copy Citation Text show less
    Flow chart of proposed algorithm
    Fig. 1. Flow chart of proposed algorithm
    Comparison of different algorithms under noise. (a) Census transform; (b) MCT; (c) proposed improved Census transform
    Fig. 2. Comparison of different algorithms under noise. (a) Census transform; (b) MCT; (c) proposed improved Census transform
    Disparity maps with different algorithms. (a) Left images; (b) true parallax maps; (c) color features; (d) gradient features; (e) Census transform; (f) proposed algorithm
    Fig. 3. Disparity maps with different algorithms. (a) Left images; (b) true parallax maps; (c) color features; (d) gradient features; (e) Census transform; (f) proposed algorithm
    Changes of PBM under different noise. (a) Salt noise; (b) Gaussian noise
    Fig. 4. Changes of PBM under different noise. (a) Salt noise; (b) Gaussian noise
    AlgorithmNo noiseSalt noiseGaussian nosie
    5%10%15%20%5σ10σ15σ20σ
    MCT4.768.4113.1719.1225.666.919.2211.2914.11
    AD-Census4.047.5911.716.0420.466.829.8412.7216.11
    Proposed algorithm4.685.878.0111.4717.866.889.2911.3113.96
    Table 1. Average percent of bad pixels under different algorithms
    AlgorithmTsukubaVenus
    NonoccAllDiscNonoccAllDisc
    SSD+MF5.237.2724.213.685.1311.80
    RT-Census5.176.2519.311.582.3714.10
    GC1.954.229.291.763.528.73
    AdaptWeight1.381.856.900.711.196.13
    Proposed algorithm1.982.268.430.280.572.79
    AlgorithmTeddyCones
    NonoccAllDiscNonoccAllDisc
    SSD+MF16.5024.7432.8410.9919.8526.21
    RT-Census7.9817.4424.154.3811.4412.27
    GC16.5025.0124.857.7318.1715.17
    AdaptWeight7.8813.3018.603.979.798.26
    Proposed algorithm5.7211.2414.943.369.289.50
    Table 2. Average percent of bad pixels under different algorithms
    Wenjie Yu, Song Ye, Yu Guo, Jian Guo. Stereo Matching Algorithm Based on Improved Census Transform and Multi-Feature Fusion[J]. Laser & Optoelectronics Progress, 2022, 59(8): 0810011
    Download Citation