基于上下文增强与多目标语义感知的电力缺陷检测算法

胡欣,阎希玥,常娅姝,程鸿亮,肖剑,罗诗伟,马亮

Context-enhanced multi-target semantic perception algorithm for power defect detection

Xin HU,Xiyue YAN,Yashu CHANG,Hongliang CHENG,Jian XIAO,Shiwei LUO,Liang MA

表 7 网络模型对比实验

Tab.7 Comparison experiments of network models

算法 P R mAP50 mAP50-95 FLOPs/ 109 Parameters/ 106 FPS/ (帧·s−1) YOLOv6(l) 0.657 0.579 0.628 0.484 391.2 110.800 70.422 YOLOv8(l) 0.705 0.655 0.706 0.526 164.9 43.622 53.763 RT-DETR 0.696 0.571 0.607 0.449 114.9 44.220 116.279 YOLOv9(c) 0.715 0.626 0.692 0.537 78.7 25.850 163.930 YOLOv10(l) 0.667 0.600 0.648 0.501 282.3 103.700 79.365 PDD-SPCIE 0.719 0.642 0.712 0.538 122.7 49.100 78.740