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

Figure 11

The comparisons of Faster R-CNN under normal, hazy, and low-light imaging conditions. F R-CNN represents the Faster R-CNN. (a) Original. (b) Original + F R-CNN. (c) Hazy. (d) Hay + F R-CNN. (e) Low-light. (f) Low-light + F R-CNN.

(a)

(b)

(c)

(d)

(e)

(f)