以多幅图像非几何约束线段匹配重建建筑物外立面三维线段模型

张琳彦 1; 朱利 2

doi:10.7652/xjtuxb201404003

您当前的位置：

首页 >

文章列表页 >

以多幅图像非几何约束线段匹配重建建筑物外立面三维线段模型

更新时间：2025-04-29

- 以多幅图像非几何约束线段匹配重建建筑物外立面三维线段模型
- 3D Line Segments Reconstruction for Building Facades with Line Matching Across Multi-Image with Non-Geometry Constraint
- 西安交通大学学报 2014年48卷第4期页码：15-19+25
- 作者机构：
  
  西安交通大学电子与信息工程学院,西安,710049
- 作者简介：
- 基金信息：
  
  : 2013-09-12。作者简介: 张琳彦(1988—),女,硕士生;朱利(通信作者),男,副教授。基金项目: 深圳市南山区技术研发和创意设计项目(KC2012JSJS0019A)
- DOI：10.7652/xjtuxb201404003
  中图分类号： TP391
- 网络首发：2014-04-10，
  
  纸质出版：2014
- 稿件说明：
移动端阅览
张琳彦 1, 朱利 2. 以多幅图像非几何约束线段匹配重建建筑物外立面三维线段模型[J]. 西安交通大学学报, 2014,48(4):15-19+25.

3D Line Segments Reconstruction for Building Facades with Line Matching Across Multi-Image with Non-Geometry Constraint[J]. 2014, 48(4): 15-19+25.
张琳彦 1, 朱利 2. 以多幅图像非几何约束线段匹配重建建筑物外立面三维线段模型[J]. 西安交通大学学报, 2014,48(4):15-19+25. DOI： 10.7652/xjtuxb201404003.

3D Line Segments Reconstruction for Building Facades with Line Matching Across Multi-Image with Non-Geometry Constraint[J]. 2014, 48(4): 15-19+25. DOI： 10.7652/xjtuxb201404003.

摘要

针对建筑物外立面三维线段模型重建的现有算法中存在多幅图像线段匹配需要提前获知对极几何约束

以及三维线段重建的过程会因为输入误差的累积而受到影响等问题

提出了一种以非几何约束重建三维线段模型的算法。该算法在多幅图像的线段匹配过程中以点线相结合的高稳定性双视图线段匹配算法为基础

通过匹配线段间的传递性形成匹配线段组

再剔除不相容匹配并合并相容匹配

最终得到准确的匹配线段结果。三维线段重建采用分段重建的方式得到每一个匹配线段组对应的三维线段

再经过对线段长度和角度的约束

剔除不构成建筑物外立面框架的空间线段

最终构成建筑物外立面的三维线段结构。实验结果表明

该算法可以有效重建出建筑物外立面的整体框架。

Abstract

The previous works on 3D line segment reconstruction of building facades require known geometrical constraints in multi-image line matching

and the line segment reconstruction is affected by accumulated input error. An improved algorithm with non-geometry constraint is proposed. In line matching across multi-image

robust two-view line matching method is chosen

line-point invariants are taken as the foundation and matched-line-segment sets are formed via transitivity of matched line segments. Incompatible matches are eliminated and compatible matches are merged from those sets to form an accurate result. For this line segments reconstruction

3D line segments corresponding to each matched-line-segment set are obtained in piecewise

and the length constraint and angle constraint are considered to remove the line segments which do not constitute building facades. The whole structure of building facades is thus effectively reconstructed with this algorithm.

关键词

Keywords

references

CHEN Tingwang, WANG Qing. 3D line segment detection for unorganized point clouds from multi-view stereo[C]∥Proceedings of the 10th Asian Conference on Computer Vision. Berlin, Germany: Springer-Verlag, 2011: 400-411.

彭亚丽, 刘芳, 刘侍刚. 一维子空间的三维重建方法[J]. 西安交通大学学报, 2009, 43(12): 31-34.

PENG Yali, LIU Fang, LIU Shigang. 3D reconstruction method with 1D subspace[J]. Journal of Xi'an Jiaotong University, 2009, 43(12): 31-34.

PARK J S. Interactive 3D reconstruction from multiple images: a primitive-based approach[J]. Pattern Recognition Letters, 2005, 26(16): 2558-2571.

von GIOI R G, JAKUBOWICZ J, MOREL J M, et al. LSD: a fast line segment detector with a false detection control[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(4): 722-732.

SCHMID C, ZISSEMAN A. Automatic line matching across views[C]∥Proceedings of the 1977 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE Computer Society, 1997: 666-671.

SCHMID C, ZISSEMAN A. The geometry and matching of lines and curves over multiple views[J]. International Journal of Computer Vision, 2000, 40(3): 199-233.

BAY H, FERRARIS V, van GOOL L. Wide-baseline stereo matching with line segments[C]∥ Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2005: 329-336.

WANG Zhiheng, WU Fuchao, HU Zhanyi. MSLD: a robust descriptor for line matching[J]. Pattern Recognition, 2009, 42(5): 941-953.

FAN Bin, WU Fuchao, HU Zhanyi. Robust line matching through line-point invariants[J]. Pattern Recognition, 2012, 45(2): 794-805.

BAILLARD C, SCHMID C, ZISSEMAN A, et al. Automatic line matching and 3D reconstruction of buildings from multiple views[C]∥ Proceedings of the ISPRS Conference on Automatic Extraction of GIS Objects from Digital Imagery. Amsterdam, Netherlands, Holland: Elsevier Science, 1999: 69-80.

WERNER T. LMATCH: Matlab toolbox for matching line segments across multiple calibrated images[EB/OL].(2007-06-01)[2013-06-12]. http:∥cmp.felk.cvut.cz/～werner/software/lmatch/.

HARTLEY R, ZISSEMAN A. Multiple view geometry in computer vision[M]. Cambridge, UK: Cambridge University Press, 2000.

WERNER T, ZISSERMAN A. New techniques for automated architectural reconstruction from photographs[C]∥ Proceedings of the 7th European Conference on Computer Vision. Berlin, Germany: Springer-Verlag, 2002: 541-555.

VERETNIK S. Benchmark data set of Wadham College[DB/OL].(2006-01-01)[2013-6-23]. http:∥www.robots.ox.ac.uk/～vgg/data/data-mview.html.

NORONHA S, NEVATIA R. Detection and modeling of buildings from multiple aerial images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(5): 501-518.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

电子级超纯氮深冷空分流程节能优化研究

采用机器学习的大规模并行程序I/O性能建模与预测方法

融合物理信息与深度算子网络的变压器绕组热-流场快速计算方法

隧道内燃料电池汽车氢气泄漏扩散特性及影响因素

三塔合一间接空冷系统结构参数对夏季运行特性的影响研究