Research Article
[Retracted] A Way to Automatically Generate Lane Level Traffic Data from Video in the Intersections
Table 4
Fine-grained lane flow data in intersection.
| ā | Lane ID | Volume in five minutes | Volume in five minutes (ground truth) | Accuracy |
| Scene 1 (Tahe road) | 1 | 25 | 32 | 0.78 | 2 | 26 | 32 | 0.81 | 3 | 23 | 29 | 0.79 | 4 | 62 | 67 | 0.93 | 5 | 49 | 54 | 0.91 | 6 | 67 | 71 | 0.94 | 7 | 20 | 24 | 0.83 | 8 | 26 | 29 | 0.9 | 9 | 68 | 75 | 0.91 | 10 | 36 | 42 | 0.86 | 11 | 34 | 39 | 0.87 |
| Scene 2 (Youjian road) | 1 | 12 | 14 | 0.86 | 2 | 22 | 26 | 0.85 | 3 | 25 | 27 | 0.93 | 4 | 16 | 19 | 0.84 | 5 | 36 | 39 | 0.92 | 6 | 74 | 76 | 0.97 | 7 | 52 | 56 | 0.93 | 8 | 43 | 46 | 0.93 | 9 | 28 | 31 | 0.9 | 10 | 36 | 37 | 0.97 |
| Scene 3 (Tianshan road) | 1 | 12 | 15 | 0.8 | 2 | 13 | 17 | 0.76 | 3 | 23 | 26 | 0.88 | 4 | 36 | 39 | 0.92 | 5 | 16 | 18 | 0.89 | 6 | 16 | 19 | 0.84 | 7 | 24 | 27 | 0.89 | 8 | 68 | 74 | 0.92 | 9 | 57 | 68 | 0.84 | 10 | 26 | 29 | 0.9 | 11 | 69 | 79 | 0.87 | 12 | 74 | 86 | 0.86 | 13 | 29 | 34 | 0.85 | 14 | 19 | 25 | 0.76 |
|
|