Laser & Optoelectronics Progress, Volume. 58, Issue 14, 1410014(2021)
Recognition Algorithm of Strip Steel Surface Defects Based on Attention Model
Figures & Tables(13)
Fig. 1. Deep residual network frame for strip steel surface defect recognition based on attention mechanism
Fig. 2. Strip surface defect dataset. (a) Cr; (b) In; (c) Pa; (d) Ps; (e) Rs; (f) Sc
Fig. 4. False color enhancement effect. (a) Before enhancement; (b) after enhancement
Fig. 6. Training curves of model under different learning rates. (a) Accuracy curves; (b) loss function curves
Fig. 7. Pictures with different signal-to-noise ratios. (a) No noise; (b) 50 dB; (c) 40 dB; (d) 30 dB; (e) 20 dB
Table 1. Distribution of strip surface defect datasets
View tableTable 1. Distribution of strip surface defect datasets
Type of defects Cr In Pa Ps Rs Sc Before expansion 300 300 300 300 300 300 After expansion 900 900 900 900 900 900
Table 2. Recognition accuracy of each defect under different learning rates
View tableTable 2. Recognition accuracy of each defect under different learning rates
Learning rate Recognition accuracy /% Training accuracy /% Training loss value Average accuracy /% Cr In Pa Ps Rs Sc 0.01 100 95.00 100 85.00 100 98.33 99.24 0.0986 96.32 0.001 100 98.33 100 96.67 98.33 98.33 99.31 0.0205 98.61 0.0001 90.00 98.33 100 88.33 100 98.33 95.58 0.1154 95.83 0.00001 98.33 96.67 100 78.33 100 96.67 81.74 0.5037 95.00
Table 3. Anti-noise ability of each model under Gaussian noise with different signal-to-noise ratios
View tableTable 3. Anti-noise ability of each model under Gaussian noise with different signal-to-noise ratios
Model Recognition accuracy /% 20 dB 30 dB 40 dB 50 dB LBP[ 7 ]22.78 63.57 69.26 75.05 MB-LBP[ 7 ]25.76 71.35 90.24 91.83 Improved MB-LBP[ 7 ]30.63 78.56 95.66 97.00 ResNet50 21.65 75.56 92.22 95.56 A-ResNet50 34.44 88.33 94.44 98.61
Table 4. Recognition ability of proposed model A-ResNet50
View tableTable 4. Recognition ability of proposed model A-ResNet50
Defect category Number of samples Ppr Recall rate F1 Cr In Pa Ps Rs Sc Cr 60 0 0 0 0 0 1.000 1.000 1.000 In 0 59 0 0 1 1 0.967 0.983 0.975 Pa 0 0 60 2 0 0 0.968 1.000 0.984 Ps 0 1 0 58 0 0 0.983 0.967 0.975 Rs 0 0 0 0 59 0 1.000 0.983 0.992 Sc 0 0 0 0 0 59 1.000 0.983 0.992
Table 5. Recognition results of strip steel surface defects by different models
View tableTable 5. Recognition results of strip steel surface defects by different models
Model Recognition accuracy /% Average accuracy /% Unit inference time /s Cr In Pa Ps Rs Sc Model in Ref.[5] 97.00 94.00 98.00 98.00 97.00 93.00 96.17 0.065 Model in Ref.[7] 99.00 98.00 97.00 96.00 97.00 95.00 97.00 0.058 AlexNet 100 98.33 100 75.00 100 98.33 95.28 0.016 A-AlexNet 100 95.00 100 85.00 100 98.33 96.39 0.017 GoogleNet 81.67 100 91.67 81.67 90.00 98.33 90.56 0.389 A-GoogleNet 93.33 100 96.67 90.00 90.00 100 95.00 0.041 ResNet50 93.33 95.00 100 85.00 100 100 95.56 0.074 A-ResNet50 100 98.33 100 96.67 98.33 98.33 98.61 0.078 ResNet101 100 98.33 100 88.33 98.33 100 97.50 0.121 A-ResNet101 100 96.67 100 91.67 100 100 98.05 0.130
Tools
Get Citation
Copy Citation Text
Yanuo Lu, Bingcai Chen, Degang Chen, Shixiang Yan, Shunping Li. Recognition Algorithm of Strip Steel Surface Defects Based on Attention Model[J]. Laser & Optoelectronics Progress, 2021, 58(14): 1410014
Paper Information
Category: Image Processing
Received: Sep. 30, 2020
Accepted: Nov. 14, 2020
Published Online: Jun. 30, 2021
The Author Email: Bingcai Chen (china@dlut.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20