一种仿射不变的直线描述子与直线匹配

doi:10.3969/j.issn.0372-2112.2015.12.024

电子学报 ›› 2015, Vol. 43 ›› Issue (12): 2505-2512.DOI: 10.3969/j.issn.0372-2112.2015.12.024

一种仿射不变的直线描述子与直线匹配

缪君^1,2, 储珺², 张桂梅²

1. 南昌大学机电工程学院, 江西南昌 330031;
2. 南昌航空大学计算机视觉研究所, 江西南昌 330063

收稿日期:2014-09-03 修回日期:2014-11-26 出版日期:2015-12-25
- 通讯作者:
- 储珺
- 作者简介:
- 缪君男,1979年生于江西崇仁,2008年获西北工业大学硕士学位,南昌大学机电工程学院博士研究生.南昌航空大学航空制造工程学院讲师.主要研究方向为计算机视觉与图像处理.E-mail:miaojun@nchu.edu.cn;张桂梅 女,1970年生于江西临川,2006年获西北工业大学博士学位,现为南昌航空大学航空制造工程学院教授.主要研究方向为图像处理,计算机视觉与模式识别等.E-mail:zhangguimei@nchu.edu.cn
- 基金资助:
- 国家自然科学基金 (No.61263046,No.61462065); 江西省自然科学基金 (No.20122BAB201037)

An Affine Invariant Line Descriptor and Line Matching

MIAO Jun^1,2, CHU Jun², ZHANG Gui-mei²

1. School of Mechanical & Electrical Engineering, Nanchang University, Nanchang, Jiangxi 330031, China;
2. Institute of Computer Vision, Nanchang Hangkong University, Nanchang, Jiangxi 330063, China

Received:2014-09-03 Revised:2014-11-26 Online:2015-12-25 Published:2015-12-25

摘要/Abstract

摘要：

从图像中提取的直线常出现不完整、端点位置不准确等问题,针对这些问题造成的直线匹配难点,本文提出了一种仿射不变的直线描述子.首先将待匹配直线离散为对应点的集合,将直线描述转化为点的描述,避免了直线不完整造成的支撑区域大小不一致的问题;然后结合直线的方向和长度,定义点描述子的主方向和尺度,通过统计离散点集的局部邻域的梯度信息使描述子具有仿射不变性.为了提高直线匹配速度,在进行直线描述之前,本文采用了极线约束精简了待匹配直线集合,再利用最近邻距离比准则对直线精确匹配.实验结果表明本文提出的直线描述子在仿射、亮度、视点、遮挡等变化条件下具有精确的匹配性能.

关键词: 直线匹配, 直线描述子, 仿射不变, 极线约束

Abstract:

Line matching is a difficult problem due to reasons such as incomplete lines,inaccurate locations of endpoints,and so on.To deal with these challenges,we proposed an affine invariant line descriptor.An initial candidate match is dispersed to a set of correspondences. As a result,the problem of inconsistent support region size is resolved because we need only construct descriptors of correspondence points instead of lines. In order to make the descriptor affine invariant,the dominant orientation and the scale of the descriptor are created according to the direction and the length of the line,and gradients of the discrete points set in the local neighborhood are calculated. To speed up line matching,epipolar constraint is used before constructing line descriptors,and the number of potential matches is limited. Then,line matching is preceded accurately by the nearest neighbor distance ratio approach. The experimental results show that the proposed descriptor has accurate line matching under the changes of affine,illumination,viewpoint,and partial occlusion.

Key words: line matching, line descriptor, affine invariant, epipolar constraint

中图分类号:

TP391.9

缪君, 储珺, 张桂梅. 一种仿射不变的直线描述子与直线匹配[J]. 电子学报, 2015, 43(12): 2505-2512.

MIAO Jun, CHU Jun, ZHANG Gui-mei. An Affine Invariant Line Descriptor and Line Matching[J]. Acta Electronica Sinica, 2015, 43(12): 2505-2512.

参考文献

[1] KOUTSOUDIS A,VIDMAR B,IOANNAKIS G,et al.Multi-image 3D reconstruction data evaluation[J].Journal of Cultural Heritage,2014,15(1):73-79.
[2] 余烨,刘晓平,BUCKLES B P.基于数据融合的居民区建筑物重建方法研究[J].电子学报,2014,42(2):251-256. YU Y,LIU X P,BUCKLES B P.Residential building reconstruction based on data fusion[J].Acta Electronica Sinica,2014,42(2):251-256.(in Chinese)
[3] HAN J,KIM D,LEE M,SUNWOO M.Enhanced road boundary and obstacle detection using a downward-looking LIDAR sensor[J].IEEE Transactions on Vehicular Technology,2012,61(3):971-985.
[4] FAN B,WU F C,HU Z Y.Robust line matching through line–point invariants[J].Pattern Recognition,2012,45(2):794-805.
[5] LIU Z, MARLET R.Virtual line descriptor and semi-local matching method for reliable feature correspondence[A].Proceedings of British Machine Vision Conference[C].Guildford,UK:British Machine Vision Association,2012.1-11.
[6] 张云生,朱庆,吴波,邹峥嵘.一种基于三角网约束的立体影像线特征多级匹配方法[J].武汉大学学报:信息科学版,2013,38(5):522-527. ZHANG Y S,ZHU Q,WU B,ZOU Z R.A hierarchical stereo line matching method based on a triangle constraint[J].Geomatics and Information Science of Wuhan University,2013,38(5):522-527.(in Chinese)
[7] SCHMID C,ZISSENMAN A.The geometry and matching of lines and curves over multiple views[J].International Journal of Computer Vision,2000,40(3):199-233.
[8] BAY H,FERRARI V,VAN G L.Wide-baseline stereo matching with line segments[A].Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C].San Diego,USA:IEEE Computer Society,2005.329-336.
[9] 王志衡,吴福朝.均值-标准差描述子与直线匹配[J].模式识别与人工智能,2009,22(1):32. WANG Z H,WU F C.Mean-standard deviation descriptor and line matching[J].Pattern Recognition and Artificial Intelligence,2009,22(1):32.(in Chinese)
[10] WANG Z H,WU F C,HU Z Y.MSLD:A robust descriptor for line matching[J].Pattern Recognition,2009,42(5):941-953.
[11] LOWE D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[12] ZHANG L,KOCH R.Line matching using appearance similarities and geometric constraints[J].Pattern Recognition.2012,7476:236-245.
[13] ZHANG L L,KOCH R:An efficient and robust line segment matching approach based on LBD descriptor and pairwise geometric consistency[J].Journal of Visual Communication and Image Representation,2013,24(7):794-805.
[14] VERHAGEN B,TIMOFTE R,VAN G L.Scale-invariant line descriptors for wide baseline matching[A].Proceedings of the IEEE Winter Conference on Applications of Computer Vision[C].Steamboat Springs,Colorado,USA:IEEE,2014.493-500.
[15] WANG L,NEUMANN U,YOU S.Wide-baseline image matching using line signatures[A].Proceedings of the IEEE International Conference on Computer Vision[C].Kyoto,Japan:IEEE,2009.1311-1318.
[16] 周焰,李德仁,徐长勇.尺度变化对形状的比例不变特征的影响[J].华中科技大学学报:自然科学版,2003,31(3):11-13. ZHOU Y,LI D R,XU C Y.The effect of scale variance on features with scale in variance[J].Journal of Huazhong University of Science and Technology (Natural Science Edition),2003,31(3):11-13.(in Chinese)
[17] CANNY J.A computational approach to edge detection[J].IEEE Transactions on Pattern Analysis and Machine Intelli- gence,1986,8(6):679-698.
[18] FAN B.Line Matching Leveraged by Point Correspondences[EB/OL].http://vision.ia.ac.cn/Students/bfan/lm.htm,2011-05-19.
[19] VERHAGEN B.Scale-Invariant Line Descriptors for Wide Baseline Matching[EB/OL].https://github.com/bverhagen/SMSLD,2014-01-25.

一种仿射不变的直线描述子与直线匹配

An Affine Invariant Line Descriptor and Line Matching

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

[1]	黄波, 赵晓晖, 庞怡杰, 时公涛, 陈东, 赵继印. 一种多尺度自卷积快速算法[J]. 电子学报, 2013, 41(12): 2430-2435.
[2]	杨建伟, 蓝如师. 仿射不变的中心投影[J]. 电子学报, 2012, 40(11): 2315-2319.
[3]	黄波;赵继印;郑蕊蕊;李敏. 基于多尺度自卷积归一化直方图的仿射不变量模式识别[J]. 电子学报, 2011, 39(1): 64-69.
[4]	蔡红苹, 雷琳, 陈涛, 粟毅. 一种通用的仿射不变特征区域提取方法[J]. 电子学报, 2008, 36(4): 672-678.
[5]	贾建华;焦李成;黄文涛. 一种基于质心不变特性的仿射不变纹理特征提取算法[J]. 电子学报, 2008, 36(10): 1910-1915.
[6]	黄旭明, 王斌, 张立明. 一种基于独立元分析的仿射不变描述和仿射变换参数估计的新方法[J]. 电子学报, 2004, 32(12): 2020-2023.
[7]	张泽旭;李金宗;李宁宁. 基于光流场分割和Canny边缘提取融合算法的运动目标检测[J]. 电子学报, 2003, 31(9): 1299-1302.