基于改进YOLOv5s的烟梗物料目标检测算法

吕佳铭,张峰,罗亚波

Improved YOLOv5s based target detection algorithm for tobacco stem material

Jiaming LV,Feng ZHANG,Yabo LUO

表 3 不同目标检测算法对比实验结果分析

Tab.3 Analysis of experimental results for comparing different object detection algorithms

模型 P/% R/% mAP@0.50/% mAP@0.50∶0.95/% M/MB GFLOPs FPS/帧 Faster R-CNN 78.6 83.3 87.5 82.0 108.0 150.8 13.14 YOLOv3-tiny 75.8 85.1 84.3 80.8 16.6 13.0 163.93 YOLOv3 78.2 87.1 87.9 83.8 117.9 155.3 27.80 YOLOv7-tiny 85.2 88.7 89.8 86.4 11.7 13.2 123.46 YOLOv7 75.3 87.7 86.3 84.8 71.3 105.2 25.58 YOLOx-s 76.5 92.9 89.1 84.0 34.3 26.8 101.73 YOLOv8s 80.3 93.5 89.5 86.7 21.4 28.8 112.36 YOLOv5s 77.8 93.3 90.3 89.0 13.8 16.6 212.77 本研究算法 86.2 94.1 96.1 94.7 12.1 21.3 178.57