基于线特征的鱼眼图像与地面激光雷达点云配准

doi:10.13474/j.cnki.11-2246.2021.0212

测绘通报 ›› 2021, Vol. 0 ›› Issue (7): 74-80.doi: 10.13474/j.cnki.11-2246.2021.0212

基于线特征的鱼眼图像与地面激光雷达点云配准

熊友谊¹, 乔纪纲², 张文金¹

1. 广州欧科信息技术股份有限公司, 广东广州 510663;
2. 广东财经大学地理与旅游学院, 广东广州 510320

收稿日期:2020-09-03 出版日期:2021-07-25 发布日期:2021-08-04
通讯作者: 乔纪纲。E-mail:qjg821@163.com
作者简介:熊友谊(1971-),男,博士,高级工程师,主要从事三维地理信息系统、激光扫描技术与全景影像技术的研究。E-mail:xiongyy@okaygis.com
基金资助:
广东省科技计划（2019B010151001）；广州市科技计划（201604010125）

Registration of fisheyes panoramic image and terrestrial LiDAR date based on line feature

XIONG Youyi¹, QIAO Jigang², ZHANG Wenjin¹

1. Guangzhou Ouke Information Technology Corporation, Guangzhou 510663, China;
2. School of Geography and Tourism, Guangdong University of Finance and Economics, Guangzhou 510320, China

Received:2020-09-03 Online:2021-07-25 Published:2021-08-04

摘要/Abstract

摘要： 地面激光扫描数据（LiDAR）与全景图像进行联合分析在虚拟场景建设、文化遗迹保护等方面有较高的应用价值。将全景图像在其视点与LiDAR点云对齐是开展两种数据联合分析的基础，但需要克服数据维度差异转换和特征匹配的困难。本文提出了基于线特征的半自动配准方法纠正图像，将鱼眼图像和LiDAR点云投影为透视成像的柱面全景图像，采用Hough变换提取图像直线特征，并利用修正迭代Hough变换方法，实现在鱼眼全景图像视点约束下与离散激光点云的三维对齐。试验表明，该方法能在较少的人工干预下实现二维到三维数据对齐。

关键词: 鱼眼图像, LiDAR, 线特征, MIHT, 配准

Abstract: Combining terrestrial laser point cloud and fisheye image are high efficiency on panoramic information analysis. Registration of fisheyes panoramic image and terrestrial LiDAR data is a prerequistite for its information complement each other. But it is necessary to overcome the difficulties of data dimensional difference transformation and feature matching. In this paper, a new semi-automatic method is brought up to registered these two data based on Hough algorithm, both fisheyes image and LiDAR points are reprojected to a same cylindrical surface. Line features are extracted from these two images based on Hough algorithm. And then, the modified Iterated Hough transform is employed to registered LiDAR range image and fisheye cylindrical surface image. Experimentation shows that the model has capability to register these two data in 3D screen with less manual intervention, and obtain high register precision within the area near fisheyes image viewing point.

Key words: fisheyes image, LiDAR, line feature, MIHT, registration

中图分类号:

P237

熊友谊, 乔纪纲, 张文金. 基于线特征的鱼眼图像与地面激光雷达点云配准[J]. 测绘通报, 2021, 0(7): 74-80.

XIONG Youyi, QIAO Jigang, ZHANG Wenjin. Registration of fisheyes panoramic image and terrestrial LiDAR date based on line feature[J]. Bulletin of Surveying and Mapping, 2021, 0(7): 74-80.

参考文献 19

[1]	AGOSTO E, BOR WAZ L. 3D models in cultural heritage:approaches for their creation and use[J]. International Journal of Computational Methods in Heritage Science, 2017,1(1):1-9.
[2]	鲁春芳. 面向全景立体球视觉系统的三维重建关键算法研究[D].天津:天津理工大学,2016.
[3]	司传波, 高子翔, 崔程宝. 机载LiDAR数据和航空影像基于角特征的配准[J].测绘与空间地理信息, 2017, 40(10):151-153.
[4]	盛庆红,张斌,肖晖,等.直线簇约束下的地面LiDAR点云配准方法[J].武汉大学学报(信息科学版),2018,43(3):406-412.
[5]	HAO W, WANG YH, LIANG W. Slice-based building facade reconstruction from 3D point clouds[J]. International Journal of Remote Sensing, 2018,39(20):6587-6606.
[6]	BARBARELLA M, D'AMICO F, BLASIIS M R D, et al. Use of terrestrial laser scanner for rigid airport pavement management.[J]. Sensors (Basel), 2017,18(1):44.
[7]	闫利,索一凡,曹亮.车载激光雷达点云与全景影像的自动配准[J].测绘科学,2016,41(4):113-117.
[8]	FUJIWARA E, SILVA L E D, MARQUES T H R,et al. Polymer optical fiber specklegram strain sensor with extended dynamic range[J].Optical Engineering,2018,57(11):116107.
[9]	陈驰,杨必胜,田茂,等.车载MMS激光点云与序列全景影像自动配准方法[J].测绘学报,2018,47(2):215-224.
[10]	皮英冬,李欣,陈智勇,等.基于三维控制场的鱼眼影像检校和纠正方法[J].光学学报,2017,37(1):224-232.
[11]	王向军,白皓月,吴凡璐,等.基于改进球面透视投影的鱼眼图像畸变校正方法[J].图学学报,2018,39(1):43-49.
[12]	张晓康,李小平.浅析模糊图像复原方法[J].现代计算机(专业版),2018(14):38-42.
[13]	THAHER R H, HUSSEIN Z K. Stereo vision distance estimation employing canny edge detector with interpolation algorithm[J]. American Journal of Science and Technology,2015,2(6):346-353.
[14]	BURGOS P B, MANSO L J, BUSTOS P. A variant of the Hough transform for the combined detection of corners, segments, and polylines[J]. EURASIP Journal on Image and Video Processing,2017,32:10.1186.
[15]	刁燕,吴晨柯,罗华,等.基于改进的概率Hough变换的直线检测优化算法[J].光学学报,2018,38(8):170-178.
[16]	MANZANERA A, NGUYEN P T, XU X L. Line and circle detection using dense one-to-one Hough transforms on greyscale images[J]. EURASIP Journal on Image and Video Processing,2016(1):10.1186.
[17]	VIANELLO A, MANFREDI G, DIEBOLD M, et al. 3D reconstruction by a combined structure tensor and Hough transform light field approach[J]. tm-Technisches Messen,2017,84,7:10.1515.
[18]	MENG Y C, ZHANG ZP, YIN HQ, et al. Automatic detection of particle size distribution by image analysis based on local adaptive canny edge detection and modified circular Hough transform.[J]. Micron,2018,106:34-41.
[19]	李喆,李建增,胡永江,等.基于频谱预处理与改进霍夫变换的离焦模糊盲复原算法[J].图学学报,2018,39(5):909-916.

基于线特征的鱼眼图像与地面激光雷达点云配准

Registration of fisheyes panoramic image and terrestrial LiDAR date based on line feature

可视化

摘要/Abstract

引用本文

使用本文

参考文献 19

相关文章 15

编辑推荐

Metrics

本文评价

[1]	王艳军, 林云浩, 王书涵, 李少春, 王孟杰. OSM辅助车载LiDAR点云三维道路边界精细提取[J]. 测绘通报, 2022, 0(7): 18-25.
[2]	温雨笑, 吕杰, 马庆勋, 张鹏, 徐汝岭. 高光谱和LiDAR联合反演森林生物量研究[J]. 测绘通报, 2022, 0(7): 38-42.
[3]	经皓然, 徐敬海. 异形建筑的无人机点云与地面三维激光扫描点云配准[J]. 测绘通报, 2022, 0(6): 32-39.
[4]	谢洪, 陈立波, 聂倩, 吴玮, 张沛. 利用点云配准的空地影像融合技术[J]. 测绘通报, 2022, 0(6): 82-87.
[5]	谢元, 董梦, 马新建. 多源数据在复杂场景建筑物三维重建的应用[J]. 测绘通报, 2022, 0(6): 143-147.
[6]	王广帅. 无人机倾斜摄影铁路轨道线高精度自动重建[J]. 测绘通报, 2022, 0(5): 133-139,156.
[7]	徐福斌, 曲相屹, 王健, 刘秀涵. 基于LiDAR点云的复杂钢结构模型重建[J]. 测绘通报, 2022, 0(5): 170-173.
[8]	阮明浩, 卢永华, 刘玉贤, 汤圣君, 王伟玺, 张宇琦, 李游. 激光与倾斜点云精确配准与无缝融合[J]. 测绘通报, 2022, 0(4): 6-10.
[9]	普东东, 丁海勇. 融合超体素与成对链接聚类的LiDAR点云分割算法[J]. 测绘通报, 2022, 0(3): 65-69.
[10]	高莎, 袁希平, 甘淑, 杨明龙, 袁新悦, 罗为东. 融合无人机影像与LiDAR点云的山区地表覆被景观特征探测[J]. 测绘通报, 2022, 0(1): 110-115.
[11]	高智梅, 王竞雪, 沈昭宇. 机载LiDAR建筑物点云渐进提取算法[J]. 测绘通报, 2021, 0(8): 7-13,36.
[12]	闫坤, 张志华, 颜鲁春. 结合正则化K-SVD和Hampel滤波的探地雷达数据重建[J]. 测绘通报, 2021, 0(8): 22-27.
[13]	王晓静, 唐超, 王勇, 侯海倩, 杨晓飞. 封闭环境下限制误差发散的高精度导航定位[J]. 测绘通报, 2021, 0(8): 65-68,73.
[14]	阎跃观, 陈中章, 孙阳, 李泽政, 姚承志. 密集低矮植被区LiDAR点云地面滤波算法[J]. 测绘通报, 2021, 0(7): 1-5.
[15]	高莎, 袁希平, 甘淑, 杨亚复, 李霞. 面向复杂山地的机载LiDAR与无人机影像融合应用[J]. 测绘通报, 2021, 0(7): 65-69.