• Acta Photonica Sinica
  • Vol. 51, Issue 2, 0210006 (2022)
Huijuan TIAN1、2、*, Jiahao ZHAI1、2, Jianxin LIU3, Jiawei LIU1、2, and Linlin DENG1、2
Author Affiliations
  • 1Tianjin Key Laboratory of Optoelectronic Detection Technology and System,School of Electrical and Electronics Engineering,Tiangong University,Tianjin 300387,China
  • 2Engineering Research Center of Ministry of Education on High Power Solid State Lighting Application System,Tianjin 300387,China
  • 3Tianjin Chengke Transmission Electromechanical Technology Co.,Ltd.,Tianjin 300384,China
  • show less
    DOI: 10.3788/gzxb20225102.0210006 Cite this Article
    Huijuan TIAN, Jiahao ZHAI, Jianxin LIU, Jiawei LIU, Linlin DENG. A Low-quality Iris Image Segmentation Algorithm Based on SRN-UNet[J]. Acta Photonica Sinica, 2022, 51(2): 0210006 Copy Citation Text show less
    Overall network structure of SRN-UNet
    Fig. 1. Overall network structure of SRN-UNet
    Network structure of SE-ResNext module
    Fig. 2. Network structure of SE-ResNext module
    The basic unit of ResNet and ResNext
    Fig. 3. The basic unit of ResNet and ResNext
    Network structure of SENet
    Fig. 4. Network structure of SENet
    Data set sample display
    Fig. 5. Data set sample display
    Part of the data set and corresponding annotations
    Fig. 6. Part of the data set and corresponding annotations
    The prediction results of different algorithms for near-infrared iris images
    Fig. 7. The prediction results of different algorithms for near-infrared iris images
    The prediction results of different algorithms for low-quality iris images
    Fig. 8. The prediction results of different algorithms for low-quality iris images
    LossMiou/%F1 score/%Precision/%
    Cross entropy loss93.8896.8497.08
    Dice+Focal95.1997.4897.82
    Table 1. Comparison of segmentation performance of different loss functions
    AlgorithmBatch=10Batch=20Batch=50
    U-Net0.909 41.749 34.414 7
    Ours0.621 01.237 43.048 3
    Table 2. Time required for network operation(s)
    AlgorithmMiou/%F1 score/%Precision/%
    U-Net90.9995.2192.44
    SegNet91.1895.3292.15
    PSPNet92.2895.9794.12
    Deeplabv392.3196.0095.73
    Ours95.1997.4897.82
    Table 3. Comparison of different iris segmentation algorithms
    U-NetResNext-UNetSENet-UNetMiou/%F1 score/%Precision/%
    91.0295.2092.51
    91.8995.4793.18
    92.0195.3093.27
    95.1997.4897.82
    Table 4. Influence of each module on whole model
    Huijuan TIAN, Jiahao ZHAI, Jianxin LIU, Jiawei LIU, Linlin DENG. A Low-quality Iris Image Segmentation Algorithm Based on SRN-UNet[J]. Acta Photonica Sinica, 2022, 51(2): 0210006
    Download Citation