Few-Shot Object Detection for Remote Sensing Images via Pseudo-Sample Generation and Feature Enhancement

Few-shot object detection (FSOD) based on fine-tuning is essential for analyzing optical remote sensing images. However, existing methods mainly focus on natural images and overlook the scale variations in remote sensing images, leading to feature confusion among foreground instances of different cl...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhaoguo Huang, Danyang Chen, Cheng Zhong
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/8/4477
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850183642601488384
author Zhaoguo Huang
Danyang Chen
Cheng Zhong
author_facet Zhaoguo Huang
Danyang Chen
Cheng Zhong
author_sort Zhaoguo Huang
collection DOAJ
description Few-shot object detection (FSOD) based on fine-tuning is essential for analyzing optical remote sensing images. However, existing methods mainly focus on natural images and overlook the scale variations in remote sensing images, leading to feature confusion among foreground instances of different classes. Additionally, since only a subset of instances are labeled in FSOD training data, the model might mistakenly treat unlabeled instances as background, leading to confusion between foreground features and background features, particularly those of novel classes. The preceding phenomenon indicates that severe feature confusion in remote sensing FSOD hampers the ability of the model to accurately classify and localize instances. To address these issues, this paper proposes a two-stage FSOD framework based on transfer learning via pseudo-sample generation and feature enhancement (PSGFE), including pseudo-sample generation module (PSGM) and feature enhancement module (FEM). The former reduces the feature confusion between foreground and background by generating pseudo-samples for unannotated background areas. The latter dynamically captures and enhances multi-scale features on the region of interest (ROI), and extracts unique core information for each class to eliminate the feature confusion among foreground instances of different classes. Our method has been validated on the optical remote sensing datasets DIOR and RSOD. It demonstrates superior performance compared to existing methods.
format Article
id doaj-art-8b5e49a8619b42fa8a7a93206fce92a9
institution OA Journals
issn 2076-3417
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-8b5e49a8619b42fa8a7a93206fce92a92025-08-20T02:17:19ZengMDPI AGApplied Sciences2076-34172025-04-01158447710.3390/app15084477Few-Shot Object Detection for Remote Sensing Images via Pseudo-Sample Generation and Feature EnhancementZhaoguo Huang0Danyang Chen1Cheng Zhong2School of Computer, Electronics and Information, Guangxi University, Nanning 530004, ChinaSchool of Computer, Electronics and Information, Guangxi University, Nanning 530004, ChinaSchool of Computer, Electronics and Information, Guangxi University, Nanning 530004, ChinaFew-shot object detection (FSOD) based on fine-tuning is essential for analyzing optical remote sensing images. However, existing methods mainly focus on natural images and overlook the scale variations in remote sensing images, leading to feature confusion among foreground instances of different classes. Additionally, since only a subset of instances are labeled in FSOD training data, the model might mistakenly treat unlabeled instances as background, leading to confusion between foreground features and background features, particularly those of novel classes. The preceding phenomenon indicates that severe feature confusion in remote sensing FSOD hampers the ability of the model to accurately classify and localize instances. To address these issues, this paper proposes a two-stage FSOD framework based on transfer learning via pseudo-sample generation and feature enhancement (PSGFE), including pseudo-sample generation module (PSGM) and feature enhancement module (FEM). The former reduces the feature confusion between foreground and background by generating pseudo-samples for unannotated background areas. The latter dynamically captures and enhances multi-scale features on the region of interest (ROI), and extracts unique core information for each class to eliminate the feature confusion among foreground instances of different classes. Our method has been validated on the optical remote sensing datasets DIOR and RSOD. It demonstrates superior performance compared to existing methods.https://www.mdpi.com/2076-3417/15/8/4477few-shot object detectionremote sensing imagespseudo-sample generationfeature enhancement
spellingShingle Zhaoguo Huang
Danyang Chen
Cheng Zhong
Few-Shot Object Detection for Remote Sensing Images via Pseudo-Sample Generation and Feature Enhancement
Applied Sciences
few-shot object detection
remote sensing images
pseudo-sample generation
feature enhancement
title Few-Shot Object Detection for Remote Sensing Images via Pseudo-Sample Generation and Feature Enhancement
title_full Few-Shot Object Detection for Remote Sensing Images via Pseudo-Sample Generation and Feature Enhancement
title_fullStr Few-Shot Object Detection for Remote Sensing Images via Pseudo-Sample Generation and Feature Enhancement
title_full_unstemmed Few-Shot Object Detection for Remote Sensing Images via Pseudo-Sample Generation and Feature Enhancement
title_short Few-Shot Object Detection for Remote Sensing Images via Pseudo-Sample Generation and Feature Enhancement
title_sort few shot object detection for remote sensing images via pseudo sample generation and feature enhancement
topic few-shot object detection
remote sensing images
pseudo-sample generation
feature enhancement
url https://www.mdpi.com/2076-3417/15/8/4477
work_keys_str_mv AT zhaoguohuang fewshotobjectdetectionforremotesensingimagesviapseudosamplegenerationandfeatureenhancement
AT danyangchen fewshotobjectdetectionforremotesensingimagesviapseudosamplegenerationandfeatureenhancement
AT chengzhong fewshotobjectdetectionforremotesensingimagesviapseudosamplegenerationandfeatureenhancement