宁夏大学学报（自然科学版中英文）2026，Vol.47Issue(1)：33-41,9.DOI:10.20176/j.cnki.nxdz.000112
基于大核选择和形状自适应的遥感图像目标检测
Target Detection in Remote Sensing Images Based on Large Kernel Selection and Shape Adaptation
摘要
Abstract
Target detection in optical remote sensing images is a key technology for the intelligent interpretation of remote sensing data.To address the challenges posed by of significant scale variations,background interfer-ence and the diversity of target shapes in remote sensing image detection,this paper proposes the Large Multi-scale Kernel(LMK)network.This network utilized large kernel convolution and a multi-scale attention mecha-nism to dynamically adjust the spatial receptive field,thereby enhancing the capture of contextual information about objects in remote sensing scenes.Furthermore,a Shape-Adaptive Selection(SAS)label allocation strat-egy was designed for target detection,which focused on the aspect ratio of the target shapes and calculated an optimal IoU threshold based on the shape information and feature distribution.To address the difficulty of target orientation and positioning in remote sensing images,this paper introduced the KFIoU loss function.Experi-mental results show that the proposed target detection model achieves accuracies of 96.73%,97.85%,and 77.26%on the HRSC 2016,UCAS-AOD,and DOTA datasets,respectively,outperforming most existing target detection algorithms.关键词
目标检测/深度学习/标签分配/多尺度注意力/大核网络Key words
target detection/deep learning/label allocation/multi-scale attention/large kernel network分类
信息技术与安全科学引用本文复制引用
赵子澳,董爱华,黄荣..基于大核选择和形状自适应的遥感图像目标检测[J].宁夏大学学报（自然科学版中英文）,2026,47(1):33-41,9.基金项目
国家自然科学基金资助项目(62001099) （62001099）
中央高校基本科研业务费专项资金资助项目(2232023D-30) （2232023D-30）
