基于Slim-YOLOv4与嵌入式设备的无人机检测

UAV Detection Based on Slim-YOLOv4 with Embedded Device

为了在资源受限的嵌入式设备上实现对无人机的实时检测,文中提出了一种基于YOLOv4的轻量化检测网络,即Slim-YOLOv4。该网络选用Ghostnet替换原YOLOv4的主干特征提取部分,将特征融合部分中的3×3卷积替换为深度可分离卷积,并对深度可分离卷积中的激活函数进行优化,以减少网络的参数量、计算量,加快网络的收敛。实验结果表明Slim-YOLOv4的准确率达到91.6%,与原YOLOv4相比损失了1.6%,但是原YOLOv4的权重文件高达250 MB。在不影响鲁棒性的前提下,Slim-YOLOv4的权重文件大小仅为42 MB,且优于Faster-RCNN模型的108 MB和Mobilenetv3模型的53 MB。新方法每秒处理的图片数量在PC上达到31.2 frames·s^(-1),在嵌入式设备上高达37.6 frames·s^(-1),证明可以将其部署到嵌入式设备上对无人机进行实时检测。

In order to realize real-time detection of UAV on resource-constrained embedded devices,this study proposes a lightweight detection network,Slim-YOLOv4,based on YOLOv4.In this network,Ghostnet is used to replace the main feature extraction part of the original YOLOv4,the 3×3 convolution in the feature fusion part is replaced by the deep detachability convolution and the activation function in the deep detachability convolution is optimized to reduce the number of parameters and computation of the network and accelerate the convergence of the network.The experimental results show that the accuracy of the Slim-YOLOv4 is 91.6%,which is 1.6%lower than that of the original YOLOv4,but the weight file of original YOLOv4 is up to 250 MB.On the premise of not affecting the robustness,the weight file size of Slim-YOLOv4 is only 42 MB,which is better than the 108 MB of Fan-RCNN model and 53 MB of Mobilenetv3 model.The number of images processed per second by the proposed method reaches 31.2 frames·s-1 on PCS and 37.6 frames·s-1 on embedded devices,proving that the proposed method can be deployed on embedded devices for real-time detection of UAVs.