Open QinHF5 opened 1 year ago
If the heatmap is not normalized, the code normalizes heatmap using the maximum anomaly score in that image. You should normalize the anomaly scores using the maximum anomaly score in the dataset in advance to avoid this issue.
The visualization script shows a heat map that identifies the location of defects. However, in the test I detect the complete hazelnut, there are also parts marked in red. How should I determine what pictures are GOOD?