Table 4 CNN Model test results.
From: Defects detection of GMAW process based on convolutional neural network algorithm
Test set | Classification | Number of pictures (sheet) | Accuracy | Test speed (s/sheet) |
|---|---|---|---|---|
Penetration | Full penetration | 362 | 0.995945 | 0.060629 |
Sag depression | 224 | 0.989011 | ||
Burn through | 133 | 0.993358 | ||
Pores | With pores | 305 | 0.971365 | 0.060836 |
No pore | 295 | 0.980151 | ||
Arc light | Strong arc interference | 1121 | 0.996725 | 0.060782 |
Weak arc interference | 560 | 0.971772 | ||
Slags | With slags | 595 | 0.985129 | 0.060645 |
Little slag | 300 | 0.970987 |
