Deep Learning-Enabled Automatic Detection of Bridges for Promoting Transportation Surveillance under Different Imaging Conditions

Figure 8

The comparisons of different competing methods for bridge detection at the presence of cooccurrences of water and land. From left to right: (a) original images and detected results by (b) Faster R-CNN, (c) YOLOv3, (d) YOLOv4, and (e) YOLOv5s, respectively.

(a)

(b)

(c)

(d)

(e)