摘要
在无人机影像配准过程中,特征点对提取数量较为庞大,配准效率较低;当特征点位于树冠区域内时,受地形、视角、光照等因素的影响,非常容易产生大量点对的错误匹配。为解决上述问题,文章提出一种基于数学形态学改进的无人机影像配准算法。首先,根据可见光波段差异植被指数(visible-band difference vegetation index,VDVI)与数学形态学原理,对树冠区域内的特征点进行剔除;然后将双向匹配算法与最近邻(k-nearest neighbors,KNN)算法相结合,对特征点进行粗匹配;最后利用渐进一致采样(progressive sample consensus,PROSAC)算法进行精匹配,进一步提高配准精度。实验结果表明,在对有树冠区域覆盖的无人机影像进行配准时,该文提出的算法能有效提高匹配率与配准精度。
In the process of unmanned aerial vehicle(UAV)image registration,the number of feature point pairs extracted is large and the registration efficiency is low.When the feature points are located in the crown area,it is very easy to produce a large number of wrong matching of point pairs due to the influence of terrain,viewing angle,illumination and other factors.In order to solve the above problems,an improved UAV image registration algorithm based on mathematical morphology is proposed.Firstly,the feature points in the crown area are removed according to the visible-band difference vegetation index(VDVI)and the principle of mathematical morphology.Then the bidirectional matching algorithm is combined with the k-nearest neighbors(KNN)method to roughly match the feature points.Finally,progressive sample consensus(PROSAC)is used for precise matching to further improve the registration accuracy.The experimental results show that the proposed algorithm can effectively improve the matching rate and registration accuracy when registering UAV images covered by tree canopy.
作者
陆可
郑伯桢
卢春盛
王中元
LU Ke;ZHENG Bozhen;LU Chunsheng;WANG Zhongyuan(Jiangsu Key Laboratory of Resources and Environmental Information Engineering, China University of Mining and Technology, Xuzhou 221116, China;School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou 221116, China;Shaoguan Land and Resources Technology Center, Shaoguan 512026, China)
出处
《合肥工业大学学报（自然科学版）》
CAS
北大核心
2021年第10期1406-1412,1419,共8页
Journal of Hefei University of Technology：Natural Science
基金
江苏省科技研发计划资助项目(BK20181361)。
关键词
无人机影像
影像配准
特征点
树冠区域
可见光波段差异植被指数(VDVI)
数学形态学
unmanned aerial vehicle(UAV)image
image registration
feature point
crown area
visible-band difference vegetation index(VDVI)
mathematical morphology
