Fig. 11
From: Post-variational classical quantum transfer learning for binary classification
Visual comparison of Ants and Bees dataset as input images under different noise conditions used to evaluate the robustness of the PVCQTL model. (a) Clean images with no noise, (b) Images with Gaussian noise at 20% intensity, and (c) Grayscale images corrupted by salt-and-pepper noise at 20%. These augmentations simulate real-world perturbations to assess how well the model generalizes under degraded conditions.
