河南理工大学学报（自然科学版）2025，Vol.44Issue(4)：40-47,8.DOI:10.16186/j.cnki.1673-9787.2024070040
基于多尺度特征融合的遥感图像小目标检测
Small-object detection in remote sensing images using multi-scale feature fusion
摘要
Abstract
Objectives Small objects in remote sensing images often lack sufficient discriminative features and are highly susceptible to interference from complex backgrounds,leading to frequent false and missed detections.To address this,a multi-scale feature fusion network is proposed to improve small-object detec-tion accuracy in remote sensing images.Methods A sparse attention-guided feature fusion module is first in-troduced into the medium-scale feature maps to enhance the network's sensitivity to small objects and sup-press background interference.Furthermore,to effectively integrate contextual information across different scales and improve localization accuracy,a multi-step dilated convolution fusion module is designed.This module applies multiple parallel convolutions with varying dilation rates to aggregate semantic information from features at multiple levels.Results Extensive experiments conducted on the NWPU VHR-10,RSOD,and HRSID datasets demonstrate that the proposed method achieves significantly improved detection accu-racy for small objects while maintaining or slightly enhancing performance on medium-and large-scale ob-jects.The mAP@50 values on the NWPU VHR-10,RSOD,and HRSID datasets reached 63.1%,96.92%,and 92.5%,respectively.Conclusions These results demonstrate that the proposed method,which incorpo-rates two multi-scale feature fusion strategies based on attention guidance and dilated convolution,can ef-fectively enhance the detection accuracy of small objects in remote sensing targets.关键词
小目标检测/多尺度/遥感图像/深度学习/特征融合Key words
small-object detection/multi scale/remote sensing image/deep learning/feature fusion分类
计算机与自动化引用本文复制引用
魏明军,葛一珲,杨轩,刘亚志,李辉..基于多尺度特征融合的遥感图像小目标检测[J].河南理工大学学报（自然科学版）,2025,44(4):40-47,8.基金项目
国家重点研发计划项目(2017YFE0135700) （2017YFE0135700）
河北省高等学校科学技术研究项目(ZD2022102) （ZD2022102）
