Dual efficient self-attention network for multi-target detection in aerial imagery

被引:1
|
作者
Wang, Sikui [1 ,3 ,4 ]
Liu, Yunpeng [1 ,2 ,4 ,5 ]
Lin, Zhiyuan [1 ,3 ,4 ]
Zhang, Zhongyu [1 ,3 ,4 ]
机构
[1] Chinese Acad Sci, Shenyang Inst Automat, Shenyang, Peoples R China
[2] Chinese Acad Sci, Inst Robot & Intelligent Mfg, Shenyang, Peoples R China
[3] Univ Chinese Acad Sci, Beijing, Peoples R China
[4] Chinese Acad Sci, Key Lab Optoelect Informat Proc, Shenyang, Peoples R China
[5] Key Lab Image Understanding & Comp Vis, Shenyang, Peoples R China
关键词
Target detection; Self-attention block; Deconvolutional module; Semantic features; Hard examples mining;
D O I
10.1117/12.2549196
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Aerial 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.
引用
收藏
页数:7
相关论文
共 50 条
  • [1] Lightweight multi-target detection algorithm for unmanned aerial vehicle aerial imagery
    Liu, Yang
    Ma, Ding
    Wang, Yongfu
    [J]. JOURNAL OF APPLIED REMOTE SENSING, 2023, 17 (04)
  • [2] Stance Detection with a Multi-Target Adversarial Attention Network
    Sun, Qingying
    Xi, Xuefeng
    Sun, Jiajun
    Wang, Zhongqing
    Xu, Huiyan
    [J]. ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING, 2023, 22 (02)
  • [3] DPNET: DUAL-PATH NETWORK FOR EFFICIENT OBJECT DETECTION WITH LIGHTWEIGHT SELF-ATTENTION
    Shi, Huimin
    Zhou, Quan
    Ni, Yinghao
    Wu, Xiaofu
    Latecki, Longin Jan
    [J]. 2022 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, 2022, : 771 - 775
  • [4] A Self-attention Network for Face Detection Based on Unmanned Aerial Vehicles
    Hua, Shunfu
    Fan, Huijie
    Ding, Naida
    Li, Wei
    Tang, Yandong
    [J]. INTELLIGENT ROBOTICS AND APPLICATIONS (ICIRA 2022), PT II, 2022, 13456 : 440 - 449
  • [5] Efficient Multi-View Graph Convolutional Network with Self-Attention for Multi-Class Motor Imagery Decoding
    Tan, Xiyue
    Wang, Dan
    Xu, Meng
    Chen, Jiaming
    Wu, Shuhan
    [J]. BIOENGINEERING-BASEL, 2024, 11 (09):
  • [6] A self-attention network for smoke detection
    Jiang, Minghua
    Zhao, Yaxin
    Yu, Feng
    Zhou, Changlong
    Peng, Tao
    [J]. FIRE SAFETY JOURNAL, 2022, 129
  • [7] An Aerial Target Recognition Algorithm Based on Self-Attention and LSTM
    Liang, Futai
    Chen, Xin
    He, Song
    Song, Zihao
    Lu, Hao
    [J]. Computers, Materials and Continua, 2024, 81 (01): : 1101 - 1121
  • [8] An Improved End-to-End Multi-Target Tracking Method Based on Transformer Self-Attention
    Hong, Yong
    Li, Deren
    Luo, Shupei
    Chen, Xin
    Yang, Yi
    Wang, Mi
    [J]. REMOTE SENSING, 2022, 14 (24)
  • [9] Multi-head enhanced self-attention network for novelty detection
    Zhang, Yingying
    Gong, Yuxin
    Zhu, Haogang
    Bai, Xiao
    Tang, Wenzhong
    [J]. PATTERN RECOGNITION, 2020, 107
  • [10] SpotNet: Self-Attention Multi-Task Network for Object Detection
    Perreault, Hughes
    Bilodeau, Guillaume-Alexandre
    Saunier, Nicolas
    Heritier, Maguelonne
    [J]. 2020 17TH CONFERENCE ON COMPUTER AND ROBOT VISION (CRV 2020), 2020, : 230 - 237