Skip to main content
. 2019 Jul 14;19(14):3111. doi: 10.3390/s19143111

Figure 5.

Figure 5

Representative detection results. (a) More foreground objects could be detected by the DualNet-300 model. (b) Some cases in which DualNet-300 performs not as well as the MobileNetV2–SSD. (c) Some challenging examples for both MobileNetV2–SSD-300 and DualNet-300.