MOON: A Subspace-Based Multi-Branch Network for Object Detection in Remotely Sensed Images

The effectiveness of training-based object detection heavily depends on the amount of sample data. But in the field of remote sensing, the amount of sample data is difficult to meet the needs of network training due to the non-cooperative imaging modes and complex imaging conditions. Moreover, the i...

Full description

Bibliographic Details
Main Authors: Huan Zhang, Wei Leng, Xiaolin Han, Weidong Sun
Format: Article
Language:English
Published: MDPI AG 2023-08-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/15/17/4201
_version_ 1827727877718671360
author Huan Zhang
Wei Leng
Xiaolin Han
Weidong Sun
author_facet Huan Zhang
Wei Leng
Xiaolin Han
Weidong Sun
author_sort Huan Zhang
collection DOAJ
description The effectiveness of training-based object detection heavily depends on the amount of sample data. But in the field of remote sensing, the amount of sample data is difficult to meet the needs of network training due to the non-cooperative imaging modes and complex imaging conditions. Moreover, the imbalance of the sample data between different categories may lead to the long-tail problem during the training. Given that similar sensors, data acquisition approaches, and data structures could make the targets in different categories possess certain similarities, those categories can be modeled together within a subspace rather than the entire space to leverage the amounts of sample data in different subspaces. To this end, a subspace-dividing strategy and a subspace-based multi-branch network is proposed for object detection in remotely sensed images. Specifically, a combination index is defined to depict this kind of similarity, a generalized category consisting of similar categories is proposed to represent the subspace, and a new subspace-based loss function is devised to address the relationship between targets in one subspace and across different subspaces to integrate the sample data from similar categories within a subspace and to balance the amounts of sample data between different subspaces. Furthermore, a subspace-based multi-branch network is constructed to ensure the subspace-aware regression. Experiments on the DOTA and HRSC2016 datasets demonstrated the superiority of our proposed method.
first_indexed 2024-03-10T23:14:11Z
format Article
id doaj.art-38cb0aea97ff4c90985d9c1a6ecaac0e
institution Directory Open Access Journal
issn 2072-4292
language English
last_indexed 2024-03-10T23:14:11Z
publishDate 2023-08-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj.art-38cb0aea97ff4c90985d9c1a6ecaac0e2023-11-19T08:45:51ZengMDPI AGRemote Sensing2072-42922023-08-011517420110.3390/rs15174201MOON: A Subspace-Based Multi-Branch Network for Object Detection in Remotely Sensed ImagesHuan Zhang0Wei Leng1Xiaolin Han2Weidong Sun3Department of Electronic Engineering, Tsinghua University, Beijing 100084, ChinaDepartment of Electronic Engineering, Tsinghua University, Beijing 100084, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, ChinaDepartment of Electronic Engineering, Tsinghua University, Beijing 100084, ChinaThe effectiveness of training-based object detection heavily depends on the amount of sample data. But in the field of remote sensing, the amount of sample data is difficult to meet the needs of network training due to the non-cooperative imaging modes and complex imaging conditions. Moreover, the imbalance of the sample data between different categories may lead to the long-tail problem during the training. Given that similar sensors, data acquisition approaches, and data structures could make the targets in different categories possess certain similarities, those categories can be modeled together within a subspace rather than the entire space to leverage the amounts of sample data in different subspaces. To this end, a subspace-dividing strategy and a subspace-based multi-branch network is proposed for object detection in remotely sensed images. Specifically, a combination index is defined to depict this kind of similarity, a generalized category consisting of similar categories is proposed to represent the subspace, and a new subspace-based loss function is devised to address the relationship between targets in one subspace and across different subspaces to integrate the sample data from similar categories within a subspace and to balance the amounts of sample data between different subspaces. Furthermore, a subspace-based multi-branch network is constructed to ensure the subspace-aware regression. Experiments on the DOTA and HRSC2016 datasets demonstrated the superiority of our proposed method.https://www.mdpi.com/2072-4292/15/17/4201object detectionlong-tail problemgeneralized categorysubspace-based multi-branch networkremotely sensed image
spellingShingle Huan Zhang
Wei Leng
Xiaolin Han
Weidong Sun
MOON: A Subspace-Based Multi-Branch Network for Object Detection in Remotely Sensed Images
Remote Sensing
object detection
long-tail problem
generalized category
subspace-based multi-branch network
remotely sensed image
title MOON: A Subspace-Based Multi-Branch Network for Object Detection in Remotely Sensed Images
title_full MOON: A Subspace-Based Multi-Branch Network for Object Detection in Remotely Sensed Images
title_fullStr MOON: A Subspace-Based Multi-Branch Network for Object Detection in Remotely Sensed Images
title_full_unstemmed MOON: A Subspace-Based Multi-Branch Network for Object Detection in Remotely Sensed Images
title_short MOON: A Subspace-Based Multi-Branch Network for Object Detection in Remotely Sensed Images
title_sort moon a subspace based multi branch network for object detection in remotely sensed images
topic object detection
long-tail problem
generalized category
subspace-based multi-branch network
remotely sensed image
url https://www.mdpi.com/2072-4292/15/17/4201
work_keys_str_mv AT huanzhang moonasubspacebasedmultibranchnetworkforobjectdetectioninremotelysensedimages
AT weileng moonasubspacebasedmultibranchnetworkforobjectdetectioninremotelysensedimages
AT xiaolinhan moonasubspacebasedmultibranchnetworkforobjectdetectioninremotelysensedimages
AT weidongsun moonasubspacebasedmultibranchnetworkforobjectdetectioninremotelysensedimages