A Deep Learning Method for Building Extraction from Remote Sensing Images by Fuzing Local and Global Features

Figure 19

Building extraction results of different networks: (a) original image; (b) ground truth; (c) Unet; (d) Segnet; (e) TransUnet; (f) MATUnet; (g) MAPNet; (h) MSRFNet; and (i) TransFuse.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)