SIA OpenIR  > 光电信息技术研究室
Dual efficient self-attention network for multi-target detection in aerial imagery
Wang SK(王思奎)1,3,4; Liu YP(刘云鹏)1,2,4,5; Lin ZY(林智远)1,3,4; Zhang ZY(张钟毓)1,3,4
Department光电信息技术研究室
Conference Name2nd Target Recognition and Artificial Intelligence Summit Forum 2019
Conference DateAugust 28-30, 2019
Conference PlaceShenyang, China
Author of SourceChinese Society for Optical Engineering
Source PublicationSecond Target Recognition and Artificial Intelligence Summit Forum
PublisherSPIE
Publication PlaceBellingham, USA
2019
Pages1-7
Indexed ByEI ; CPCI(ISTP)
EI Accession number20201108282820
WOS IDWOS:000546230500012
Contribution Rank1
ISSN0277-786X
ISBN978-1-5106-3631-6
KeywordTarget detection Self-attention block Deconvolutional module Semantic features Hard examples mining
AbstractAerial imagery target detection has been widely used in the military and economic fields. However, it still faces a variety of challenges. In this paper, we proposed several efficiency improvements based on YOLO v3 framework for getting a better small target detection precision. Firstly, a dual self-attention (DAN) block is embedded in Darknet-53's ResNet units to refine the feature map adaptively. Furthermore, the deep semantic features are cascaded with the shallow outline features in a feedforward deconvolutional module to obtain context details of small targets. Finally, introducing online hard examples mining and combining Focal Loss to enhance the discriminating ability between classes. The experimental results on the VEDAI aerial dataset show that the proposed algorithm is significantly improved in accuracy compared to the original network and achieves better performance than two-stage algorithms.
Language英语
Citation statistics
Document Type会议论文
Identifierhttp://ir.sia.cn/handle/173321/26421
Collection光电信息技术研究室
Corresponding AuthorWang SK(王思奎)
Affiliation1.Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
2.Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, China
3.University of Chinese Academy of Sciences, Beijing, China
4.Key Laboratory of Opto-Electronic Information Processing, CAS, Shenyang, China
5.Key Lab of Image Understanding and Computer Vision, Shenyang, China
Recommended Citation
GB/T 7714
Wang SK,Liu YP,Lin ZY,et al. Dual efficient self-attention network for multi-target detection in aerial imagery[C]//Chinese Society for Optical Engineering. Bellingham, USA:SPIE,2019:1-7.
Files in This Item:
File Name/Size DocType Version Access License
Dual efficient self-(860KB)会议论文 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wang SK(王思奎)]'s Articles
[Liu YP(刘云鹏)]'s Articles
[Lin ZY(林智远)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wang SK(王思奎)]'s Articles
[Liu YP(刘云鹏)]'s Articles
[Lin ZY(林智远)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wang SK(王思奎)]'s Articles
[Liu YP(刘云鹏)]'s Articles
[Lin ZY(林智远)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Dual efficient self-attention network for multi-target detection in aerial imagery.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.