Dong Zhuo, Junfeng Jing, Huanhuan Zhang, Zebin Su. Classification of Chopped Strand Mat Defects Based on Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2019, 56(10): 101009

Search by keywords or author
- Laser & Optoelectronics Progress
- Vol. 56, Issue 10, 101009 (2019)

Fig. 1. Architecture of convolutional neural network model

Fig. 2. Training process of convolutional neural network

Fig. 3. Flow chart of experiment

Fig. 4. Number of defect samples

Fig. 5. Dataset samples. (a)-(d) Parallel; (e)-(h) poor dispersion; (i)-(l) yarn knot; (m)-(p) stain

Fig. 6. Comparison of initialization parameters. (a) Training accuracy; (b) verification accuracy; (c) training loss values; (d) verification loss values

Fig. 7. Training process of fine-tuning. (a) Training accuracy; (b) verification accuracy; (c) training loss values; (d) verification loss values
|
Table 1. Comparison of initialization parameters
|
Table 2. Model performances
|
Table 3. Confusion matrix of test sample
|
Table 4. Evaluation of network performances

Set citation alerts for the article
Please enter your email address