elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

SCAF-Net: Scene Context Attention-Based Fusion Network for Vehicle Detection in Aerial Imagery

Wang, Minghui and Li, Qingpeng and Gu, Yunchao and Fang, Leyuan and Zhu, Xiao Xiang (2022) SCAF-Net: Scene Context Attention-Based Fusion Network for Vehicle Detection in Aerial Imagery. IEEE Geoscience and Remote Sensing Letters, 19, p. 3508305. IEEE - Institute of Electrical and Electronics Engineers. doi: 10.1109/LGRS.2021.3107281. ISSN 1545-598X.

[img] PDF - Only accessible within DLR bis January 2023 - Postprint version (accepted manuscript)
10MB

Official URL: https://ieeexplore.ieee.org/abstract/document/9530866

Abstract

In recent years, deep learning methods have achieved great success for vehicle detection tasks in aerial imagery. However, most existing methods focus only on extracting latent vehicle target features, and rarely consider the scene context as vital prior knowledge. In this letter, we propose a scene context attention-based fusion network (SCAF-Net), to fuse the scene context of vehicles into an end-to-end vehicle detection network. First, we propose a novel strategy, patch cover, to keep the original target and scene context information in raw aerial images of a large scale as much as possible. Next, we use an improved YOLO-v3 network as one branch of SCAF-Net, to generate vehicle candidates on each patch. Here, a novel branch for the scene context is utilized to extract the latent scene context of vehicles on each patch without any extra annotations. Then, these two branches above are concatenated together as a fusion network, and we apply an attention-based model to further extract vehicle candidates of each local scene. Finally, all vehicle candidates of different patches, are merged by global nonmax suppress (g-NMS) to output the detection result of the whole original image. Experimental results demonstrate that our proposed method outperforms the comparison methods with both high detection accuracy and speed. Our code is released at https://github.com/minghuicode/SCAF-Net.

Item URL in elib:https://elib.dlr.de/145686/
Document Type:Article
Title:SCAF-Net: Scene Context Attention-Based Fusion Network for Vehicle Detection in Aerial Imagery
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Wang, MinghuiState Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang UniversityUNSPECIFIED
Li, QingpengState Laboratory of Robot Visual Perception and Control Technology, School of Robotics, Hunan Universityhttps://orcid.org/0000-0002-7401-0928
Gu, YunchaoState Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang UniversityUNSPECIFIED
Fang, LeyuanState Laboratory of Robot Visual Perception and Control Technology, School of Robotics, Hunan Universityhttps://orcid.org/0000-0003-2351-4461
Zhu, Xiao XiangGerman Aerospace Center, Remote Sensing Technology Institutehttps://orcid.org/0000-0001-5530-3613
Date:2022
Journal or Publication Title:IEEE Geoscience and Remote Sensing Letters
Refereed publication:Yes
Open Access:No
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:Yes
Volume:19
DOI :10.1109/LGRS.2021.3107281
Page Range:p. 3508305
Publisher:IEEE - Institute of Electrical and Electronics Engineers
ISSN:1545-598X
Status:Published
Keywords:Attention-based model, deep learning, fusionnetwork, remote sensing, vehicle detection
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Earth Observation
DLR - Research theme (Project):R - Artificial Intelligence
Location: Oberpfaffenhofen
Institutes and Institutions:Remote Sensing Technology Institute > EO Data Science
Deposited By: Rösel, Anja
Deposited On:17 Nov 2021 14:42
Last Modified:20 Dec 2021 18:06

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.