• Laser & Optoelectronics Progress
  • Vol. 58, Issue 2, 0210005 (2021)
Zheng Zhang1 and Yang Xu1、2、*
Author Affiliations
  • 1College of Big Data and Information Engineering, Guizhou University, Guiyang, Guizhou 550025, China
  • 2Guiyang Aluminum Magnesium Design & Research Institute Co., Ltd., Guiyang, Guizhou 550009, China;
  • show less
    DOI: 10.3788/LOP202158.0210005 Cite this Article Set citation alerts
    Zheng Zhang, Yang Xu. Adaptive One-Hand and Two-Hand Gesture Recognition Based on Double Classifiers[J]. Laser & Optoelectronics Progress, 2021, 58(2): 0210005 Copy Citation Text show less
    Maximum pooling example
    Fig. 1. Maximum pooling example
    Classification process for jth category
    Fig. 2. Classification process for jth category
    One-hand and double-hand gesture recognition structure based on two classifiers
    Fig. 3. One-hand and double-hand gesture recognition structure based on two classifiers
    Network structure of hand number classifier
    Fig. 4. Network structure of hand number classifier
    Calculation of distance between centers of gavity of hand gestures
    Fig. 5. Calculation of distance between centers of gavity of hand gestures
    Diagrams of gesture grouping prediction. (a) Gesture binary graphs; (b) centers of gravity of hand gestures; (c) gesture grouping prediction maps
    Fig. 6. Diagrams of gesture grouping prediction. (a) Gesture binary graphs; (b) centers of gravity of hand gestures; (c) gesture grouping prediction maps
    Adaptive enhanced convolutional neural network structure
    Fig. 7. Adaptive enhanced convolutional neural network structure
    Nine types of gesture samples from ASL
    Fig. 8. Nine types of gesture samples from ASL
    Samples of one-hand and double-hand gesture data sets. (a) One-hand gestures; (b) double-hand gestures
    Fig. 9. Samples of one-hand and double-hand gesture data sets. (a) One-hand gestures; (b) double-hand gestures
    Data expansion and complex background gesture samples. (a) Complex background gestures; (b) data expansion
    Fig. 10. Data expansion and complex background gesture samples. (a) Complex background gestures; (b) data expansion
    Convergence and error rate curves of CNN and AE-CNN. (a) Convergence curves of CNN, CNN+Dropout,and AE-CNN; (b) error rate curves of CNN and AE-CNN
    Fig. 11. Convergence and error rate curves of CNN and AE-CNN. (a) Convergence curves of CNN, CNN+Dropout,and AE-CNN; (b) error rate curves of CNN and AE-CNN
    LBP features of hand gestures (0,2,5, and 9). (a) LBP feature of zero gesture; (b) LBP feature of two gesture; (c) LBP feature of five gesture; (d) LBP feature of nine gesture
    Fig. 12. LBP features of hand gestures (0,2,5, and 9). (a) LBP feature of zero gesture; (b) LBP feature of two gesture; (c) LBP feature of five gesture; (d) LBP feature of nine gesture
    HOG features of partial gestures and HOG+PCA dimensionality reduction reconstruction maps
    Fig. 13. HOG features of partial gestures and HOG+PCA dimensionality reduction reconstruction maps
    Preprocessing graphs after adding different noise. (a) Normalization of salt and pepper noise; (b) binary map of salt and pepper noise; (c) binary map of Gaussian noise; (d) distribution of Gaussian noise density; (e) normalization of Gaussian noise
    Fig. 14. Preprocessing graphs after adding different noise. (a) Normalization of salt and pepper noise; (b) binary map of salt and pepper noise; (c) binary map of Gaussian noise; (d) distribution of Gaussian noise density; (e) normalization of Gaussian noise
    NameConvolution kernel
    C13×3(32)
    S12×2 max pooling
    C23×3(64)
    S22×2 max pooling
    Dropout0.5
    Table 1. Network parameters of hand number classifier
    ClassifierCA
    Training set1021021100
    Test set19203960
    Table 2. Amount of data of classification networks
    MethodRecognition rate /%
    LBP+SVM[15]89.73
    HOG+SVM[16]91.81
    PCA+HOG+SVM[17]94.35
    AE-CNN97.87
    Table 3. Comparison of recognition rate between AE-CNN and other algorithms
    NoiseGaussian noiseSalt and pepper noise
    00.0010.0020.00300.0010.0020.003
    Recognition rate /%97.1096.8496.4996.0497.1096.7796.6196.32
    Table 4. Comparison of recognition rate between Gaussian noise and salt and pepper noise
    GroupNumber of images with complex backgroundNumber of images under different lighting conditionsRecognition rate /%
    1401095.26
    2311993.65
    3351594.37
    4282293.41
    5242693.34
    Table 5. Recognition rate of samples under complex background and different lighting conditions
    Zheng Zhang, Yang Xu. Adaptive One-Hand and Two-Hand Gesture Recognition Based on Double Classifiers[J]. Laser & Optoelectronics Progress, 2021, 58(2): 0210005
    Download Citation