基于改进YOLOv8的船舶目标检测算法
|
|
朵琳,殷瑜,段威,张芸,任勇
|
Ship target detection algorithm based on improved YOLOv8
|
|
Lin DUO,Yu YIN,Wei DUAN,Yun ZHANG,Yong REN
|
|
| 表 3 SSDD 数据集的不同目标检测模型的比较 |
| Tab.3 Comparison of different object detection models in SSDD dataset |
|
| 方法 | p/% | r/% | mAP50/% | Np/106 | FLOPs/109 | | FBR-Net[15] | 92.4 | 93.1 | 94.2 | 31.30 | 29.4 | | TWC-Net[16] | 91.2 | 95.1 | 94.1 | 26.36 | 20.8 | | DCMSNN[17] | 90.3 | 83.5 | 90.3 | 20.70 | 21.6 | | TOOD[18] | 83.1 | 93.1 | 97.1 | 72.23 | 38.8 | | Key-Point Estimation[19] | 94.8 | 95.1 | 97 .7 | 73.30 | 49.6 | | ADERLNet-CW[20] | 98.1 | 95.4 | 98.3 | 38.20 | 105.2 | | DD-YOLO | 98.5 | 93.3 | 98.5 | 13.94 | 9.8 |
|
|
|