Research Article
Research on Information Fusion of Computer Vision and Radar Signals in UAV Target Identification
Table 1
Experiment 1: identification of UAVs by the information fusion model in a good light and short distance scenario.
| Group number | 1 | 2 |
| UAV type | Helicopter | 4-rotor drone | Helicopter | 4-rotor drone | Coordinate in the image | (745, 559) | (1197, 581) | (775, 886) | (1153, 567) | Recognition accuracy by YOLOv3 | 0.96 | 0.90 | 0.85 | 0.84 | Coordinate by radar | (2.5, 3.3, 2.9) | (−3.7, 3.1, 2.1) | (2.9, 3.4, 2.6) | (−3.5, 3.2, 1.9) | Transformed coordinate | (743, 562) | (1199, 577) | (775, 887) | (1150, 566) | Recognition accuracy by radar | 0.91 | 0.86 | 0.93 | 0.92 | Recognition accuracy by the proposed system | 0.96 | 0.90 | 0.93 | 0.92 | Improved accuracy | 5.5% | 4.7% | 9.4% | 9.5% |
|
|