• Laser & Optoelectronics Progress
  • Vol. 57, Issue 6, 061503 (2020)
Zaifeng Shi1、*, Peng Ye1, Cheng Sun1, Tao Luo2, Hanjie Wang3, and Huizhuo Pan3
Author Affiliations
  • 1School of Microelectronics, Tianjin University, Tianjin 300072, China
  • 2College of Intelligence and Computing, Tianjin University, Tianjin 300072, China
  • 3School of Life Sciences, Tianjin University, Tianjin 300072, China
  • show less
    DOI: 10.3788/LOP57.061503 Cite this Article Set citation alerts
    Zaifeng Shi, Peng Ye, Cheng Sun, Tao Luo, Hanjie Wang, Huizhuo Pan. Object Detection Algorithm Applied to Optical Genetic Laser Projection System[J]. Laser & Optoelectronics Progress, 2020, 57(6): 061503 Copy Citation Text show less
    Schematic of the laser projection system
    Fig. 1. Schematic of the laser projection system
    Network structure diagram
    Fig. 2. Network structure diagram
    Packet convolution. (a) Operation flow; (b) channel shuffle
    Fig. 3. Packet convolution. (a) Operation flow; (b) channel shuffle
    Part of the mouse image dataset
    Fig. 4. Part of the mouse image dataset
    mAP at different γ and αt values
    Fig. 5. mAP at different γ and αt values
    Recall and precision results of different algorithms on mouse image dataset
    Fig. 6. Recall and precision results of different algorithms on mouse image dataset
    Comparison of mouse image detection with different algorithms. (a)(e) Faster RCNN; (b)(f) SSD; (c)(g) YOLOv3; (d)(h) proposed YOLOv3
    Fig. 7. Comparison of mouse image detection with different algorithms. (a)(e) Faster RCNN; (b)(f) SSD; (c)(g) YOLOv3; (d)(h) proposed YOLOv3
    MethodPASCAL VOC2007Mouse image dataset
    mAP /%Speed /msmAP /%Speed /ms
    Faster RCNN[7]73.220091.4208
    SSD[9]74.36179.364
    YOLOv3[12]79.22988.432
    YOLOv3 (proposed)79.82590.326
    Table 1. mAP and speed results of different algorithms
    Zaifeng Shi, Peng Ye, Cheng Sun, Tao Luo, Hanjie Wang, Huizhuo Pan. Object Detection Algorithm Applied to Optical Genetic Laser Projection System[J]. Laser & Optoelectronics Progress, 2020, 57(6): 061503
    Download Citation