Research Article
A Method for Identifying Defects on Highly Reflective Roller Surface Based on Image Library Matching
Table 1
Comparison of the results of the proposed method and several other methods.
| Category | Traditional Otsu method | Adaptive canny edge identification method [9] | The proposed method |
| Number of complete recognition | 39 | 41 | 53 | Defect recognition rate | 72.2% | 75.9% | 98.2% | Average time (s) | 0.13 | 0.53 | 0.25 |
|
|