摘要
针对公路隧道采集的衬砌裂缝图像存在重叠和间隙两种现象,不能保证裂缝首尾相连而直接拼接的问题,本文提出一种基于空间拓扑和形态相似关系的长裂缝拼接算法.首先将空间拓扑关系与图像点特征相结合,提出一种基于RCC的长裂缝连接性特征点识别算法,自动识别相邻图像中的裂缝是否同属于一条长裂缝以及存在重叠还是间隙现象;然后,采用基于余弦相似性度量的裂缝配准算法完成重叠区域的配准拼接,采用几何标定法补足间隙裂缝后首尾相连完成拼接.该算法在工程应用中得以验证,能够在图库中自动完成长裂缝的拼接.实验结果表明,在拼接精度和速度方面的综合性价比优于特征拼接算法和几何拼接算法.另外,该算法给多幅图像中存在连续性线条时,如何进行拼接提供了一种新思路.
In view of the fact that the lining crack images collected from highway tunnels have overlaps and gaps,which cannot guarantee that the cracks are connected end to end and directly splicing,this paper proposes a long crack splicing algorithm based on spatial topology and morphological similarity.First,the spatial topological relationship is combined with the image point features,and a long crack connectivity feature point recognition algorithm based on RCC is proposed to automatically identify whether the cracks in adjacent images belong to the same long crack and whether there is overlap or gap;then,a crack registration algorithm based on the cosine similarity measure is used to complete the registration and splicing of the overlapping area,and the geometric calibration method is used to fill the gap and the cracks are connected end to end to complete the splicing.The algorithm has been verified in engineering applications and can automatically complete the splicing of long cracks in the gallery.The experimental results show that the integrated cost performance is superior to the feature stitching algorithm and geometric stitching algorithm in terms of precision and speed.In addition,the algorithm provides a new idea on how to splice continuous lines in multiple images.
作者
薛丹
苑玮琦
XUE Dan;YUAN Wei-qi(Computer Vision Group,Shenyang University of Technology,Shenyang,Liaoning 110870,China)
出处
《电子学报》
EI
CAS
CSCD
北大核心
2021年第4期672-678,共7页
Acta Electronica Sinica
