YoloV8改进策略：Neck层改进、注意力改进|HCANet全局与局部的注意力模块CAFM|二次创新|即插即用
yolov9-c summary: 620 layers, 52330674 parameters, 0 gradients, 245.5 GFLOPsClass Images Instances P R mAP50 mAP50-95: 100%|██████████| 15/15 00:06all 230 1412 0.917 0.985 0.99 0.735c17 230 131 0.981 0.992 0.995 0.829c5 230 68 0.898 1 0.994 0.841helicopter 230 43 0.964 1 0.975 0.648c130 230 85 0.969 1 0.995 0.7f16 230 57 0.887 0.965 0.982 0.689b2 230 2 0.757 1 0.995 0.649other 230 86 0.953 0.94 0.963 0.557b52 230 70 0.97 0.971 0.983 0.826kc10 230 62 0.969 0.984 0.988 0.832command 230 40 0.96 1 0.995 0.815f15 230 123 0.978 1 0.995 0.672kc135 230 91 0.979 0.989 0.983 0.72a10 230 27 0.966 0.963 0.974 0.48b1 230 20 0.967 1 0.995 0.677aew 230 25 0.916 1 0.98 0.792f22 230 17 0.895 1 0.995 0.738p3 230 105 0.985 1 0.995 0.794p8 230 1 0.585 1 0.995 0.597f35 230 32 0.969 0.987 0.993 0.574f18 230 125 0.972 0.992 0.987 0.815v22 230 41 0.966 1 0.995 0.711su-27 230 31 0.979 1 0.995 0.849il-38 230 27 0.962 1 0.995 0.804tu-134 230 1 0.583 1 0.995 0.895su-33 230 2 1 0.821 0.995 0.697an-70 230 2 0.757 1 0.995 0.823tu-22 230 98 0.984 1 0.995 0.809
Results saved to runs\train\exp7