Acta Photonica Sinica, Volume. 52, Issue 8, 0817001(2023)
AtG-DeepLab V3+ Endoscopic Image Enhancement Algorithm Based on Self-attention Mechanism Optimization
Figures & Tables(9)
Fig. 1. AtG-DeepLab V3+ model structure based on self-attention mechanism and DeepLab V3+ network
Fig. 2. Optical endoscopic imaging system diagram based on fiber bundle
Fig. 4. Comparison of image resolution enhancement by AtG-DeepLab V3+ model
Fig. 5. Comparison of evaluation indicators of test target images enhanced by each model
Fig. 6. Comparison of biological tissue images enhanced by each model
Fig. 7. Comparison of evaluation indicators of biological tissue image enhanced by each model
Table 1. Model parameter configuration
View tableView in ArticleTable 1. Model parameter configuration
Parameter Value Learning rate 0.001 Beta_1 0.9 Beta_2 0.999 Momentum 0.9 Batch size 2 Decay 0 Epochs 200
Table 2. Model training parameter index
View tableView in ArticleTable 2. Model training parameter index
Model GPU occupancy/MB Training time/（h/200 epoch） Parameter quantity/MB Reconstruction rate/ ） U2-Net 31 869 11 516 0.034 2 Attention U-Net 32 155 10.3 103 0.012 8 GARNN 19 752 17 22 0.013 7 DeepLab V3+ 14 122 9 696 0.033 0 AtG-DeepLab V3+ 14 570 9.2 701 0.031 6
Tools
Get Citation
Copy Citation Text
Jiajun CHEN, Kaixiang LI, Renjian LI, Chunlei SHAO, Guiye LI, Lingling CHEN. AtG-DeepLab V3+ Endoscopic Image Enhancement Algorithm Based on Self-attention Mechanism Optimization[J]. Acta Photonica Sinica, 2023, 52(8): 0817001
Paper Information
Category:
Received: Jan. 11, 2023
Accepted: Mar. 10, 2023
Published Online: Sep. 26, 2023
The Author Email: CHEN Lingling (chenlingling@sztu.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20