基于立体视觉和图像映射的路面病害全局定位  

Global Localization of Cement Concrete Pavement Disease Based on Stereo Vision and Image Mapping

在线阅读下载全文

作  者:高明星[1] 张文琪 马东梅 GAO Mingxing;ZHANG Wenqi;MA Dongmei(College of Energy and Transportation Engineering,Inner Mongolia Agricultural University,Hohhot 010018,China;Inner Mongolia Jiaoke Road and Bridge Construction CO.Ltd.,Hohhot 010070,China)

机构地区:[1]内蒙古农业大学能源与交通工程学院,呼和浩特010018 [2]内蒙古交科路桥建设有限公司,呼和浩特010070

出  处:《内蒙古农业大学学报(自然科学版)》2025年第1期57-65,共9页Journal of Inner Mongolia Agricultural University(Natural Science Edition)

基  金:内蒙古自治区关键技术攻关计划项目(2021GG0178)。

摘  要:随着全国道路里程的增加,基于图像的自动化道路检测成为研究热点。现有图像病害检测方法存在全局位置信息不明确的问题,本文提出了融合图像映射和单目立体视觉实现全局路面病害定位方法。首先进行病害图像分类与提取,同时实施图像单目三维重建得到点云模型;其次通过图像映射,以转换矩阵和缩放因子获得现实尺度的三维病害点云;在此基础上聚类病害点云,得到单一病害参数及全局位置信息;最后将位置信息与地面真值进行验证和对比。结果表明,所提方法与地面真值的平均相对误差为0.66%、平均绝对误差为2.45 cm。With the increase of road mileage in China,image-based automated road detection has become a research hotspot.However,existing methods for detecting pavement distress often lack precise global location information.This paper addressed this limitation by proposing a novel approach for accurately locating pavement diseases on a global scale.The proposed method leveraged the image mapping and monocular stereo vision techniques.Initially,the disease images were classified and extracted,followed by the genera-tion of a point cloud model through monocular 3D reconstruction.Subsequently,image mapping was employed,utilizing the transfor-mation matrix and scaling factor calculations to obtain a realistic-scale 3D disease point cloud.Building upon this foundation,clus-tering techniques were utilized to extract individual disease parameters and global location information from the disease point clouds.Finally,the position information was verified and compared against the ground truth.The experimental results showed that the pro-posed method had an average relative error of 0.66%and an average absolute error of 2.45 cm when compared to the ground truth.

关 键 词:点云处理 三维重建 三维映射 定位 距离测量 

分 类 号:P215[天文地球—测绘科学与技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象