FRORS: An Effective Fine-Grained Retrieval Framework for Optical Remote Sensing Images

Fine-grained retrieval of remote sensing images is an image interpretation task that is still in its infancy. With the rapid development of convolutional neural networks (CNN) in the field of remote sensing, it has become possible for remote sensing image retrieval tasks to move toward more fine-gra...

Full description

Saved in:
Bibliographic Details
Main Authors: Yong-Qiang Mao, Zhizhuo Jiang, Yu Liu, Yiming Zhang, Kehan Qi, Hanbo Bi, You He
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10904305/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849344633204637696
author Yong-Qiang Mao
Zhizhuo Jiang
Yu Liu
Yiming Zhang
Kehan Qi
Hanbo Bi
You He
author_facet Yong-Qiang Mao
Zhizhuo Jiang
Yu Liu
Yiming Zhang
Kehan Qi
Hanbo Bi
You He
author_sort Yong-Qiang Mao
collection DOAJ
description Fine-grained retrieval of remote sensing images is an image interpretation task that is still in its infancy. With the rapid development of convolutional neural networks (CNN) in the field of remote sensing, it has become possible for remote sensing image retrieval tasks to move toward more fine-grained classes. However, since current methods focus on how to construct similarity metrics between sample pairs, the model ignores the learning of fine-grained intraclass heterogeneity and interclass commonality features, which poses a huge challenge to fine-grained retrieval. To solve this problem, we propose a novel fine-grained retrieval framework of optical remote sensing (FRORS) images, which aims to improve fine-grained retrieval capabilities by constructing interaction and matching between intraclass heterogeneity features, interclass commonality features, and image features. Specifically, we first construct a fine-grained prototype memory (FPM) module, and continuously update the local prototype storage unit through a lightweight CNN to achieve a refined representation of fine-grained heterogeneity features. Furthermore, to learn interclass commonality, we propose a gram learning (GraL) strategy, which is achieved by learning the correlation between feature dimensions. On this basis, we introduce a gram-based metric match (GMM) mechanism, which fuses the prototype features representing intraclass heterogeneity and the gram vector representing interclass commonality through an embedding manner, thereby achieving the purpose of fully interactive matching between image features and fine-grained class features. With FPM, GraL, and GMM, our FRORS can better learn deep features representing fine-grained classes and promote the improvement of the network&#x0027;s fine-grained retrieval ability. Extensive experiments conducted on a self-constructed THUFG-OPT dataset prove that the proposed FRORS achieves state-of-the-art fine-grained retrieval performance, which is 5.75% higher than the baseline method on <inline-formula><tex-math notation="LaTeX">$\mathrm{mAP@10}$</tex-math></inline-formula>.
format Article
id doaj-art-55b5f33edc2946908e0a52329f7eecf5
institution Kabale University
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-55b5f33edc2946908e0a52329f7eecf52025-08-20T03:42:37ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01187406741910.1109/JSTARS.2025.354582810904305FRORS: An Effective Fine-Grained Retrieval Framework for Optical Remote Sensing ImagesYong-Qiang Mao0https://orcid.org/0000-0001-9256-3668Zhizhuo Jiang1https://orcid.org/0000-0002-5269-2753Yu Liu2https://orcid.org/0000-0002-5216-3181Yiming Zhang3Kehan Qi4https://orcid.org/0009-0009-3344-3784Hanbo Bi5https://orcid.org/0009-0001-4209-5461You He6https://orcid.org/0000-0002-6111-340XDepartment of Electronic Engineering, Tsinghua University, Beijing, ChinaShenzhen International Graduate School, Tsinghua University, Shenzhen, ChinaDepartment of Electronic Engineering, Tsinghua University, Beijing, ChinaDepartment of Electronic Engineering, Tsinghua University, Beijing, ChinaShenzhen International Graduate School, Tsinghua University, Shenzhen, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaDepartment of Electronic Engineering, Tsinghua University, Beijing, ChinaFine-grained retrieval of remote sensing images is an image interpretation task that is still in its infancy. With the rapid development of convolutional neural networks (CNN) in the field of remote sensing, it has become possible for remote sensing image retrieval tasks to move toward more fine-grained classes. However, since current methods focus on how to construct similarity metrics between sample pairs, the model ignores the learning of fine-grained intraclass heterogeneity and interclass commonality features, which poses a huge challenge to fine-grained retrieval. To solve this problem, we propose a novel fine-grained retrieval framework of optical remote sensing (FRORS) images, which aims to improve fine-grained retrieval capabilities by constructing interaction and matching between intraclass heterogeneity features, interclass commonality features, and image features. Specifically, we first construct a fine-grained prototype memory (FPM) module, and continuously update the local prototype storage unit through a lightweight CNN to achieve a refined representation of fine-grained heterogeneity features. Furthermore, to learn interclass commonality, we propose a gram learning (GraL) strategy, which is achieved by learning the correlation between feature dimensions. On this basis, we introduce a gram-based metric match (GMM) mechanism, which fuses the prototype features representing intraclass heterogeneity and the gram vector representing interclass commonality through an embedding manner, thereby achieving the purpose of fully interactive matching between image features and fine-grained class features. With FPM, GraL, and GMM, our FRORS can better learn deep features representing fine-grained classes and promote the improvement of the network&#x0027;s fine-grained retrieval ability. Extensive experiments conducted on a self-constructed THUFG-OPT dataset prove that the proposed FRORS achieves state-of-the-art fine-grained retrieval performance, which is 5.75% higher than the baseline method on <inline-formula><tex-math notation="LaTeX">$\mathrm{mAP@10}$</tex-math></inline-formula>.https://ieeexplore.ieee.org/document/10904305/Fine-grainedimage retrievaloptical imagesremote sensing
spellingShingle Yong-Qiang Mao
Zhizhuo Jiang
Yu Liu
Yiming Zhang
Kehan Qi
Hanbo Bi
You He
FRORS: An Effective Fine-Grained Retrieval Framework for Optical Remote Sensing Images
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Fine-grained
image retrieval
optical images
remote sensing
title FRORS: An Effective Fine-Grained Retrieval Framework for Optical Remote Sensing Images
title_full FRORS: An Effective Fine-Grained Retrieval Framework for Optical Remote Sensing Images
title_fullStr FRORS: An Effective Fine-Grained Retrieval Framework for Optical Remote Sensing Images
title_full_unstemmed FRORS: An Effective Fine-Grained Retrieval Framework for Optical Remote Sensing Images
title_short FRORS: An Effective Fine-Grained Retrieval Framework for Optical Remote Sensing Images
title_sort frors an effective fine grained retrieval framework for optical remote sensing images
topic Fine-grained
image retrieval
optical images
remote sensing
url https://ieeexplore.ieee.org/document/10904305/
work_keys_str_mv AT yongqiangmao frorsaneffectivefinegrainedretrievalframeworkforopticalremotesensingimages
AT zhizhuojiang frorsaneffectivefinegrainedretrievalframeworkforopticalremotesensingimages
AT yuliu frorsaneffectivefinegrainedretrievalframeworkforopticalremotesensingimages
AT yimingzhang frorsaneffectivefinegrainedretrievalframeworkforopticalremotesensingimages
AT kehanqi frorsaneffectivefinegrainedretrievalframeworkforopticalremotesensingimages
AT hanbobi frorsaneffectivefinegrainedretrievalframeworkforopticalremotesensingimages
AT youhe frorsaneffectivefinegrainedretrievalframeworkforopticalremotesensingimages