Fig. 9.

Example class attention maps of (a) images in DFC15 dataset with respect to (b) impervious, (c) water, (d) clutter, (e) vegetation, (f) building, (g) tree, (h) boat, and (i) car. Red indicates strong activations, while blue represents non-activations. Besides, normalization is performed based on each row for a fair comparison among class attention maps of the same images. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)