Bag of Visual Words Model with Deep Spatial Features for Geographical Scene Classification

Figure 8

The architecture of the proposed CNN.