An Open Access Journal
Table 5 Track counting results
From: Vision-based vehicle detection and counting system using deep learning in highway scenes
Scenes | Scene 1 | Scene 2 | Scene 3 | Direction correct rate | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
Video frames | 11000 | 22500 | 41000 | |||||||||
Vehicle category | Car | Bus | Truck | Car | Bus | Truck | Car | Bus | Truck | |||
Direction A | Our method | 29 | 21 | 3 | 110 | 40 | 21 | 287 | 141 | 22 | ||
Actual number of vehicles | 32 | 21 | 3 | 117 | 43 | 22 | 297 | 150 | 24 | |||
Extra Number | 3 | 0 | 0 | 8 | 3 | 2 | 15 | 13 | 3 | 0.92 | ||
Missing number | 0 | 0 | 0 | 1 | 0 | 1 | 5 | 4 | 1 | |||
Correct rate | 0.906 | 1 | 1 | 0.923 | 0.930 | 0.864 | 0.933 | 0.887 | 0.833 | |||
Direction B | Our method | 41 | 37 | 4 | 117 | 69 | 13 | 300 | 168 | 15 | ||
Actual number of vehicles | 43 | 38 | 4 | 125 | 77 | 13 | 311 | 172 | 17 | |||
Extra Number | 2 | 2 | 0 | 11 | 10 | 0 | 15 | 8 | 2 | 0.931 | ||
Missing number | 0 | 1 | 0 | 3 | 2 | 0 | 4 | 4 | 0 | |||
Correct rate | 0.953 | 0.947 | 1 | 0.888 | 0.844 | 1 | 0.939 | 0.930 | 0.882 | |||
Real time rate | 1.27 | 1.35 | 1.48 | |||||||||
Average correct rate | 0.967 | 0.911 | 0.917 | 0.932 | ||||||||