结合深度可分离卷积的多源遥感融合影像目标检测

陈江浩,杨军

Object detection for multi-source remote sensing fused images based on depthwise separable convolution

Jianghao CHEN,Jun YANG

表 2 本研究算法在VEDAI数据集上与其他主流算法的目标检测结果对比

Tab.2 Comparison of object detection results between proposed algorithm and state of arts on VEDAI dataset

类别 mAP50/% GFLOPs/106 Params/106 FPS/Hz YOLOrs[26] 58.97 46.4 20.2 23.9 YOLOfusion[27] 78.60 27.3 12.5 18.2 SuperYOLO[16] 79.49 4.8 16.6 12.7 MF-YOLO[17] 76.62 — — — ICAfusion[21] 80.80 58.2 120.2 28.4 本研究算法 82.80 60.7 139.3 30.2