新疆大学学报（自然科学版中英文）2025，Vol.42Issue(4)：457-468,12.DOI:10.13568/j.cnki.651094.651316.2024.12.20.0003
公路场景下无人机遥感图像变化检测研究
Research on Change Detection of UAV Remote Sensing Image in Highway Scenes
摘要
Abstract
To meet the specific requirements of highway scenes,a dataset named ROR-CD forroad oil replen-ishment change detection and a new change detection model named SIBCD are constructed.The model reduces the number of parameters through a single feature extraction branch.A lightweight GFFM is utilized to integrate global and local information,thereby enhancing the model's adaptability to multi-scale data.A multi-scale fea-ture fusion module based on spatial and channel attention,named CFFM,is adopted to fuse multi-temporal and multi-level features,improving the model's detection performance.A series of experiments are conducted on two publicly available change detection datasets,LEVIR-CD and SYSU-CD,as well as the self-built road oil replen-ishment change detection dataset.The results demonstrate that the model achieves improved detection accuracy on multiple datasets while maintaining computational cost,thus meeting the practical requirements of highway scenes.关键词
遥感图像/变化检测/公路场景Key words
remote sensing images/change detection/highway scene分类
信息技术与安全科学引用本文复制引用
武爱英,伊克沙尼·帕依祖拉木,张一欣,王晓莉..公路场景下无人机遥感图像变化检测研究[J].新疆大学学报（自然科学版中英文）,2025,42(4):457-468,12.基金项目
国家自然科学基金"私有云跨虚拟机内容安全审核机制研究"(62162059). （62162059）
