• Chinese Journal of Lasers
  • Vol. 51, Issue 10, 1002319 (2024)
Kunpeng Tan1, Jiafeng Tang1, Zhibin Zhao1、*, Chenxi Wang1, Xingwu Zhang1, Weifeng He2, and Xuefeng Chen1
Author Affiliations
  • 1Institute of Aero-Engine, School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an 710049, Shaanxi, China
  • 2National Key Lab of Aerospace Power System and Plasma Technology, Air Force Engineering University, Xi’an 710038, Shaanxi, China
  • show less
    DOI: 10.3788/CJL240430 Cite this Article Set citation alerts
    Kunpeng Tan, Jiafeng Tang, Zhibin Zhao, Chenxi Wang, Xingwu Zhang, Weifeng He, Xuefeng Chen. Powder‑Spreading Defect Detection in Laser Powder Bed Fusion Based on Large Vision Model[J]. Chinese Journal of Lasers, 2024, 51(10): 1002319 Copy Citation Text show less
    Structure of SAM[16]
    Fig. 1. Structure of SAM[16]
    Limitation of SAM
    Fig. 2. Limitation of SAM
    Overall structure of PSAM
    Fig. 3. Overall structure of PSAM
    Position and structure of Adapter layer
    Fig. 4. Position and structure of Adapter layer
    Generating process of category mask
    Fig. 5. Generating process of category mask
    Structure of auto-prompt generator
    Fig. 6. Structure of auto-prompt generator
    Powder-spreading defects and annotation example. (a) Super-elevation; (b) incompletion; (c) hopping; (d) streaking; (e) lattice; (f) pixel-wise annotation example
    Fig. 7. Powder-spreading defects and annotation example. (a) Super-elevation; (b) incompletion; (c) hopping; (d) streaking; (e) lattice; (f) pixel-wise annotation example
    Comparison of segmentation results from different networks
    Fig. 8. Comparison of segmentation results from different networks
    NetworkmIoU /%IoU for per category /%
    BackgroundSuper-elevationIncompletionHoppingStreakingLattice
    Deeplab v356.50±0.2397.85±0.7536.74±2.4049.38±0.8932.05±2.8144.71±2.6178.26±1.48
    U-Net59.71±0.8698.61±0.0849.38±0.6160.06±0.0130.36±1.7242.56±2.2077.24±1.77
    PSAM65.02±0.4098.66±0.0350.10±0.6569.97±1.5142.51±1.6151.26±1.1477.65±1.24
    Table 1. Performance comparison of PSAM with different segmentation networks
    GroupAdapterAuto-promptmIoU /%Improvement
    153.69±0.28
    257.23±0.75+3.54 percentage points
    358.79±1.23+5.10 percentage points
    465.02±0.40+11.33 percentage points
    Table 2. Ablation experimental results
    Kunpeng Tan, Jiafeng Tang, Zhibin Zhao, Chenxi Wang, Xingwu Zhang, Weifeng He, Xuefeng Chen. Powder‑Spreading Defect Detection in Laser Powder Bed Fusion Based on Large Vision Model[J]. Chinese Journal of Lasers, 2024, 51(10): 1002319
    Download Citation