Fig. 10: The confusion matrix of DMCE-Net with VGG-VD19 on the AHE database.

This figure presents the confusion matrix of the proposed DMCE-Net using VGG-VD19 as the backbone network on the AHE database. The results are averaged over 50 random splits to ensure statistical stability and robustness. The matrix illustrates the model’s classification accuracy across all categories, where diagonal elements represent correct predictions and off-diagonal elements indicate misclassifications.