• Laser & Optoelectronics Progress
  • Vol. 59, Issue 12, 1215014 (2022)
Yiping Liu, Mingquan Zhou, Jiaojiao Kou, Yuehua Yu, Linqi Hai, Kang Li, and Haibo Zhang*
Author Affiliations
  • School of Information Science and Technology, Northwest University, Xi’an 710127, Shaanxi , China
  • show less
    DOI: 10.3788/LOP202259.1215014 Cite this Article Set citation alerts
    Yiping Liu, Mingquan Zhou, Jiaojiao Kou, Yuehua Yu, Linqi Hai, Kang Li, Haibo Zhang. Denoising of Cultural Relics Point Cloud Model Based on Unsupervised Network Framework[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1215014 Copy Citation Text show less
    Network structure diagram
    Fig. 1. Network structure diagram
    Network structure of first layer
    Fig. 2. Network structure of first layer
    Network structure of second layer
    Fig. 3. Network structure of second layer
    Point cloud model diagrams used in experiment. (a) Bunny; (b) Dragon; (c) Drill; (d) front view of terracotta model; (e) side view of terracotta model; (f) top view of terracotta model
    Fig. 4. Point cloud model diagrams used in experiment. (a) Bunny; (b) Dragon; (c) Drill; (d) front view of terracotta model; (e) side view of terracotta model; (f) top view of terracotta model
    Clean point cloud data. (a) Bunny; (b) Dragon; (c) Drill
    Fig. 5. Clean point cloud data. (a) Bunny; (b) Dragon; (c) Drill
    Cloud data with noise to be removed. (a) Bunny; (b) Dragon; (c) Drill
    Fig. 6. Cloud data with noise to be removed. (a) Bunny; (b) Dragon; (c) Drill
    Denoising results of algorithm in reference [15]. (a) Bunny; (b) Dragon; (c) Drill
    Fig. 7. Denoising results of algorithm in reference [15]. (a) Bunny; (b) Dragon; (c) Drill
    Denoising results of algorithm in reference [19]. (a) Bunny; (b) Dragon; (c) Drill
    Fig. 8. Denoising results of algorithm in reference [19]. (a) Bunny; (b) Dragon; (c) Drill
    Denoising results of algorithm in reference [17]. (a) Bunny; (b) Dragon; (c) Drill
    Fig. 9. Denoising results of algorithm in reference [17]. (a) Bunny; (b) Dragon; (c) Drill
    Denoising results of algorithm in reference [1]. (a) Bunny; (b) Dragon; (c) Drill
    Fig. 10. Denoising results of algorithm in reference [1]. (a) Bunny; (b) Dragon; (c) Drill
    Denoising results of proposed algorithm. (a) Bunny; (b) Dragon; (c) Drill
    Fig. 11. Denoising results of proposed algorithm. (a) Bunny; (b) Dragon; (c) Drill
    Comparison of fingers before and after denoising. (a) Fingers to be denoised; (b) finger denoising result
    Fig. 12. Comparison of fingers before and after denoising. (a) Fingers to be denoised; (b) finger denoising result
    Comparison of head before and after denoising. (a) Head to be denoised; (b) head denoising result
    Fig. 13. Comparison of head before and after denoising. (a) Head to be denoised; (b) head denoising result
    Comparison of upper body before and after denoising. (a) Upper body to be denoised;(b) upper body denoising result
    Fig. 14. Comparison of upper body before and after denoising. (a) Upper body to be denoised;(b) upper body denoising result
    Comparison of neckline before and after denoising. (a) Neckline to be denoised; (b) neckline denoising result
    Fig. 15. Comparison of neckline before and after denoising. (a) Neckline to be denoised; (b) neckline denoising result
    Comparison of hands before and after denoising. (a) Hands to be denoised; (b) hands denoising result
    Fig. 16. Comparison of hands before and after denoising. (a) Hands to be denoised; (b) hands denoising result
    Comparison of head before and after denoising. (a) Head to be denoised; (b) head denoising result
    Fig. 17. Comparison of head before and after denoising. (a) Head to be denoised; (b) head denoising result
    Comparison of upper body before and after denoising. (a) Upper body to be denoised; (b) local part of upper body to be denoised; (c) upper body denoising result; (d) local part of upper body denoising result
    Fig. 18. Comparison of upper body before and after denoising. (a) Upper body to be denoised; (b) local part of upper body to be denoised; (c) upper body denoising result; (d) local part of upper body denoising result
    Comparison of feet before and after denoising. (a) Feet to be denoised; (b) local part of feet to be denoised; (c) feet denoising result; (d) local part of feet denoising result
    Fig. 19. Comparison of feet before and after denoising. (a) Feet to be denoised; (b) local part of feet to be denoised; (c) feet denoising result; (d) local part of feet denoising result
    ModelBounding box diagonal percentage /%Chamfer distance /×10-8
    Algorithm in[15Algorithm in[17Algorithm in[19Algorithm in[1Proposed algorithm
    Bunny0.030.99880.93790.95541.00570.8869
    0.051.53481.53091.52711.54011.5216
    0.072.03762.00031.99572.10031.9942
    Dragon0.032.20792.19792.16932.21032.1359
    0.053.22653.21433.22313.23873.2192
    0.074.69764.69534.68954.71654.6584
    Drill0.030000.00050
    0.050.67850.67390.66890.70320.6529
    0.072.30232.29842.28862.31092.1895
    Table 1. Comparison of denoising effects of different algorithms under chamfer distance as evaluation standard
    ModelBounding box diagonal percentage /%MSE /×10-8
    Algorithm in[15Algorithm in[17Algorithm in[19Algorithm in[1Proposed algorithm
    Bunny0.031.01391.01271.01351.02911.0118
    0.051.90461.90451.90021.91081.8996
    0.072.53192.53042.53052.54222.5297
    Dragon0.030.79210.79180.78290.80340.7801
    0.051.211.19341.18851.241.1769
    0.071.63841.63791.63251.70311.6238
    Drill0.030000.0050
    0.050.250.24750.24960.26610.2427
    0.070.40960.39680.40020.44270.3954
    Table 2. Comparison of denoising effects of different algorithms under MSE as evaluation standard
    Yiping Liu, Mingquan Zhou, Jiaojiao Kou, Yuehua Yu, Linqi Hai, Kang Li, Haibo Zhang. Denoising of Cultural Relics Point Cloud Model Based on Unsupervised Network Framework[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1215014
    Download Citation