Lightweight Object Detection Networks for UAV Aerial Images Based on YOLO

扫码查看

原文链接

万方数据

外文摘要：Existing high-precision object detection algorithms for UAV(unmanned aerial vehicle)aerial images often have a large number of parameters and heavy weight,which makes it difficult to be applied to mobile devices.We propose three YOLO-based lightweight object detection networks for UAVs,named YOLO-L,YOLO-S,and YOLO-M,respectively.In YOLO-L,we adopt a deconvolution approach to explore suitable upsampling rules during training to improve the detection accuracy.The convolution-batch normalization-SiLU activation function(CBS)structure is replaced with Ghost CBS to reduce the number of parameters and weight,meanwhile Maxpool max-imum pooling operation is proposed to replace the CBS structure to avoid generating parameters and weight.YOLO-S greatly reduces the weight of the network by directly introducing CSPGhostNeck residual structures,so that the parameters and weight are respectively decreased by about 15％at the expense of 2.4％mAP.And YOLO-M adopts the CSPGhostNeck residual structure and deconvolution to reduce parameters by 5.6％and weight by 5.7％,while mAP only by 1.8％.The results show that the three lightweight detection networks proposed in this paper have good performance in UAV aerial image object detection task.

外文关键词：

Aerial imagesObject detectionDeep learningYou only look onceLightweight network

作者：

Yanshan LI、Jiarong WANG、Kunhua ZHANG、Jiawei YI、Miaomiao WEI、Lirong ZHENG、Weixin XIE

展开 >

作者单位：

ATR National Key Laboratory of Defense Technology,Shenzhen University,Shenzhen 518000,China

Guangdong Key Laboratory of Intelligent Information Processing,Shenzhen University,Shenzhen 518000,China

基金：

National Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaShenzhen Science and Technology ProjectShenzhen Science and Technology Projectother projectsother projects

项目编号：

617713196207616561871154JCYJ20180507182259896202008261540220012020KCXTD004WDZC20195500201

出版年：

2024

DOI：

10.23919/cje.2022.00.300

电子学报(英文)

CSTPCDEI

ISSN：1022-4653

年,卷(期)：2024.33(4)