Figure 7
An illustration shows six line graphs arranged in a two by three grid, all plotting A U C/A C C (Accuracy and Area Under the Curve) versus Channel Index, and labeled a through f.

Soft classification performance of six exemplary semantic categories on noisy images (σ = 0.01), (a)–(c) uses the 2 × 2 × 3 filter bank so that the x-axis has 12 channels, (d)–(f) uses the 3 × 3 × 3 filter bank so that the x-axis has 27 channels. Each subfigure shows four performance curves: train set AUC (red dashed line), train set ACC (red line), validation set AUC (green dashed line), and test set ACC (blue line).

or Create an Account

Close Modal
Close Modal