Research Article
Normalizing Flow-Based Industrial Complex Background Anomaly Detection
Table 4
The Recall_Ano score (%) on the HSRBD datasets reflects the anomaly detection rate when compared to previous methods.
| | Classes | GANomaly | SPADE | PaDiM | DifferNet | CS-Flow | CSA-Flow (ours) |
| | Scenario 1 | 65.40 | 3.85 | 80.77 | 80.77 | 61.54 | 84.62 (4.76%) | | Scenario 2 | 60.00 | 60.00 | 90.00 | 90.00 | 90.00 | 90.00 (0.00%) | | Scenario 3 | 87.50 | 75.00 | 87.50 | 75.00 | 87.50 | 87.50 (0.00%) | | Scenario 4 | 73.30 | 90.91 | 72.73 | 18.18 | 63.64 | 81.82 (−10.00%) | | Average | 71.55 | 57.44 | 82.75 | 65.99 | 75.67 | 85.98 (3.91%) |
|
|
Bold values denote the best result in the class.
|
