期刊文献+

一种基于特征表达的无人机影像匹配像对提取方法 被引量:5

Method for Extracting Matching Image Pairs of Unmanned Aerial Vehicle Images Based on Feature Expression
下载PDF
导出
摘要 针对无人机(unmanned aerial vehicle,UAV)影像三维重建中匹配像对提取适应性差、效率低、需准确的先验知识等问题,提出一种基于特征表达的无人机影像匹配像对提取方法。首先利用SiftGPU算法对一组无人机影像进行尺度不变特征转换(scale-invariant feature transform,SIFT)特征提取。其次,对SIFT特征向量集合构建分层词汇树,并分别计算各个影像与影像集中各影像的综合权重因子并按得分大小进行排序。最后,自适应计算影像检索阈值,得到最终的影像匹配像对。通过对不同地形地貌的无人机影像进行试验。实验结果表明:所提方法能快速、有效获取查询影像的匹配像对。与常规词汇树检索方法相比,查准率提高了19%~24%,查全率提高了17.4%~30.9%。尤其针对海量的无人机影像数据,所提方法具有更高的处理效率。 In order to solve the problems of poor adaptability,low efficiency and accurate prior knowledge in matching image pair extraction in 3D reconstruction of unmanned aerial vehicle(UAV)images.A matching image pair extraction method based on feature expression was proposed.Firstly,SiftGPU algorithm was used to extract scale-invariant feature transform(SIFT)features from a group of UAV images.Secondly,a hierarchical vocabulary tree was constructed for SIFT feature vector set,and the comprehensive weight factors of each image and each image in the image set were calculated and sorted according to the score size.Finally,the image retrieval threshold was calculated adaptively to obtain the final image matching pair.Through the experiments of UAV images with different topography and landforms,the proposed method can quickly and effectively obtain the matching image pairs of the query images.Compared with the conventional vocabulary tree retrieval method,the precision rate increases by 19%~24%,and the recall rate is increased by 17.4%~30.9%.Especially for massive UAV images data,the proposed method has higher processing efficiency.
作者 杨帅 任超锋 赵丽华 YANG Shuai;REN Chao-feng;ZHAO Li-hua(School of Geology Engineering and Geomatics, Chang’an University, Xi’an 710054, China)
出处 《科学技术与工程》 北大核心 2021年第24期10140-10147,共8页 Science Technology and Engineering
基金 国家自然科学基金(41801383) 国家重点研发项目(2018YFC1504805)。
关键词 匹配像对 无人机(UAV) 词汇树 影像检索 image pairs unmanned aerial vehicle(UAV) vocabulary tree image retrieval
  • 相关文献

参考文献8

二级参考文献108

共引文献925

同被引文献107

引证文献5

二级引证文献18

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部