改进YOLOv8s的轻量级无人机航拍小目标检测算法

翟亚红,陈雅玲,徐龙艳,龚玉

Improved YOLOv8s lightweight small target detection algorithm of UAV aerial image

Yahong ZHAI,Yaling CHEN,Longyan XU,Yu GONG

表 2 不同算法在VisDrone 2019数据集上的对比实验结果

Tab.2 Comparison of result of different algorithms on VisDrone 2019 dataset

模型 AP/% mAP50/% pedestrian people bicycle car van truck tricycle awning-tricycle bus motor RetinaNet 28.6 20.3 9.8 73.2 33.4 31.8 15.5 14.3 58.0 25.3 31.4 Faster R-CNN[19] 22.2 14.8 7.6 54.6 31.5 21.6 14.8 8.6 34.9 21.4 23.2 YOLOv3-LITE[20] 34.5 23.4 7.9 70.8 31.3 21.9 15.2 6.2 40.9 32.7 28.5 YOLOv5n 32.6 26.1 6.9 69.0 28.1 23.7 15.5 8.9 36.4 32.1 27.9 YOLOv5s 40.0 32.1 12.6 73.9 36.8 32.9 22.0 12.8 47.5 39.2 35.0 TPH-YOLOv5[4] 29.0 16.7 15.7 68.9 49.8 45.1 27.3 24.7 61.8 30.9 36.9 YOLOv7-tiny[21] 48.3 40.3 12.8 82.4 42.3 32.9 23.3 13.6 56.6 49.2 40.2 YOLOv8n 39.5 38.5 28.5 9.2 43.3 34.1 31.7 26.0 47.1 40.5 33.8 YOLOv8s 41.6 32.2 13.5 79.3 45.0 36.6 28.3 15.9 54.2 43.4 38.8 SPE_ YOLOv8s[22] 43.3 31.5 18.9 82.7 46.9 43.1 25.6 23.8 62.3 42.5 42.1 PVswin-YOLOv8s[23] 45.9 35.7 16.4 81.5 49.1 42.4 32.8 17.7 62.9 48.2 43.3 YOLOv9t 36.2 22.0 10.9 71.7 44.1 44.6 21.2 18.4 60.8 33.3 36.2 YOLOv10s 41.1 24.6 16.1 74.9 48.4 51.8 24.5 21.8 64.1 39.8 40.7 RTA-YOLOv8s 52.1 42.5 19.0 84.5 48.9 39.0 31.2 19.7 58.6 53.2 44.9