基于改进RT-DETR的复杂天气下铁路异物实时检测算法
牛宏侠,冯鼎超,侯涛

Real-time detection algorithm for railway foreign objects in complex weather conditions based on improved RT-DETR
Hongxia NIU,Dingchao FENG,Tao HOU
表 5 不同目标检测算法对比实验结果
Tab.5 Comparison of results from different object detection algorithms
方法mAP@0.5/%mAP@0.5:0.95/%ParametersFLOPs/109FPS/(帧·s−1)
RT-DETR-r1894.02(±0.3)76.05(±0.3)1988074857.075
SSD81.130.12)64.230.15)26578932.655
Faster R-CNN90.170.11)69.580.14)28304605908.914
Sparse R-CNN89.800.08)67.460.06)7783467723.248
YOLOv8m92.4474.582590874679.3101
YOLOv8l92.87(±0.1)74.84(±0.1)43691578168.068
YOLOv11m92.11(±0.1)76.01(±0.1)2000157967.794
YOLOv11l92.99(±0.1)75.03(±0.1)2541459787.064
TOOD[24]88.08(±0.1)70.64(±0.1)32001587199.023
DEIM[25]93.58(±0.2)76.06(±0.2)1926488756.781
文献[26]92.8874.581984754963.1121
本研究算法95.670.19)79.45(±0.3)726463256.888