基于三视图约束的基础矩阵估计

doi:10.11772/j.issn.1001-9081.2014.10.2930

计算机应用 ›› 2014, Vol. 34 ›› Issue (10): 2930-2933.DOI: 10.11772/j.issn.1001-9081.2014.10.2930

基于三视图约束的基础矩阵估计

李聪,赵红蕊,傅罡

清华大学地球空间信息研究所，北京 100084

收稿日期:2014-04-15 修回日期:2014-06-19 出版日期:2014-10-01 发布日期:2014-10-30
通讯作者: 赵红蕊
作者简介:李聪(1988-)，男，山东潍坊人，硕士研究生，主要研究方向：摄影测量与遥感图像处理；
赵红蕊（1969-），女，河北唐山人，教授，博士，主要研究方向：定量遥感与3S技术；
傅罡（1984-），男，陕西宝鸡人，博士研究生，主要研究方向：数字图像处理、遥感信息提取。
基金资助:
国家863计划项目

Fundamental matrix estimation based on three-view constraint

LI Cong,ZHAO Hongrui,FU Gang

Institute of Geomatics, Tsinghua University, Beijing 100084, China

Received:2014-04-15 Revised:2014-06-19 Online:2014-10-01 Published:2014-10-30
Contact: ZHAO Hongrui

摘要/Abstract

摘要：

考虑到只依赖对极几何关系的匹配点余差并不能完全区分匹配点的正确与否，从而影响内点集选取的情况，提出基于三视图约束的基础矩阵估计算法。首先，使用传统随机抽样一致性（RANSAC）算法计算三视图的任意两对相邻图像间的基础矩阵，确定三视图中共有的匹配点对，并计算估计基础矩阵时非共用图像上的匹配点在共用图像上的极线；然后，计算两条极线的交点与共用图像上对应匹配点间的距离，以距离值的大小作为内点判断的依据，得到新的内点集。在新内点集的基础上，采用M估计算法重新计算基础矩阵。实验结果表明：该方法可以同时降低噪声和错误匹配对基础矩阵精确计算的影响，精度优于传统鲁棒性算法，使点到极线的距离限制在0.3个像素左右，而且计算结果具有稳定性，可以被广泛地应用到基于图像序列的三维重建和摄影测量等领域中。

Abstract:

The matching points cant be decided absolutely by its residuals just relying on epipolar geometry residuals, which influences the selection of optimum inlier set. So a novel fundamental matrix calculation algorithm was proposed based on three-view constraint. Firstly, the initial fundamental matrices were estimated by traditional RANdom SAmple Consensus (RANSAC) method. Then matching points existed in every view were selected, and the epipolar lines of points not in the common view were calculated in fundamental matrix estimation. Distances between the points in common view and the intersection of its matching points epipolar lines were calculated. Under judgment based on the distances, a new optimum inlier set was obtained. Finally, the M-Estimators (ME) algorithm was used to calculate the fundamental matrices based on the new optimum inlier set. Through a mass of experiments in case of mismatching and noise, the results indicate that the algorithm can effectively reduce the influence of mismatch and noise on accurate calculation of fundamental matrices. It gets better accuracy than traditional robust algorithms by limiting distance between point and epipolar line to about 0.3 pixels, in addition, an improvement in stability. So, it can be widely applied to fields such as 3D reconstruction based on image sequence and photogrammetry.

中图分类号:

TP391.41

李聪赵红蕊傅罡. 基于三视图约束的基础矩阵估计[J]. 计算机应用, 2014, 34(10): 2930-2933.

LI Cong ZHAO Hongrui FU Gang. Fundamental matrix estimation based on three-view constraint[J]. Journal of Computer Applications, 2014, 34(10): 2930-2933.

参考文献

［1］ARMANGUE X, SALVI J. Overall view regarding fundamental matrix estimation ［J］. Image and Vision Computing, 2003,21(2):200-205.
［2］HARTLEY R. Projective reconstruction and invariants from multiple images ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1994,16(10):1036-1040.［3］HARTLEY R. In defense of the eight-point algorithm ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997,19(6):580-593.
［4］LUONG Q T, FAUGERAS O. Determining the fundamental matrix with planes: instability and new algorithms ［C］// Proceeding of the 1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 1993:489-494.
［5］ZHONG H. Research on methods for estimating the fundamental matrix ［D］. Changchun: Jilin University, 2005.(钟慧湘.基本矩阵计算方法的研究［D］.长春:吉林大学,2005.)
［6］TORR P, MURRAY D. The development and comparison of robust methods for estimating the fundamental matrix ［J］. International Journal of Computer Vision, 1997,24(3):271-300.
［7］LU S, LEI Y, KONG W, et al.Fundamental matrix estimation based on probability analysis and sampling consensus ［J］. Control and Decision, 2012,27(3):425-430.（鲁珊,雷英杰,孔韦韦,等.基于概率抽样一致性的基础矩阵估计算法［J］. 控制与决策,2012,27(3):425-430.）
［8］CARRO A I, MORROS J R. Promeds: an adaptive robust fundamental matrix estimation approach ［C］// Proceedings of the 2012 3DTV-Conference: the True Vision—Capture, Transmission and Display of 3D Video. Piscataway: IEEE Press, 2012:1-4.
［9］TORR P. Bayesian model estimation and selection for epipolar geometry and generic manifold fitting ［J］. International Journal of Computer Vision, 2002,50(1):35-61.
［10］LI Y, VELIPASALAR S, GURSOY M C. An improved evolutionary algorithm for fundamental matrix estimation ［C］// Proceedings of the 10th IEEE International Conference on Advanced Video and Signal Based Surveillance. Piscataway: IEEE Press, 2013:226-231.
［11］CHAN K-H, TANG C-Y, WU Y-L, et al.Robust orthogonal particle swarm optimization for estimating the fundamental matrix ［C］// Proceedings of the 2011 IEEE Visual Communications and Image Processing. Piscataway: IEEE Press, 2011:1-4.
［12］SHI X, LIU F, WANG Y, et al.A fundamental matrix estimation algorithm based on point weighting strategy ［C］// Proceedings of the 2011 International Conference on Virtual Reality and Visualization. Piscataway: IEEE Press, 2011:24-29.
［13］CALDERON D B, MARIA T. An approach for estimating the fundamental matrix ［C］// Proceedings of the 2011 6th Colombian Computing Congress. Piscataway: IEEE Press, 2011:1-6.
［14］BRANDT S S. Maximum likelihood robust regression by mixture models ［J］. Journal of Mathematical Imaging and Vision, 2006,25(1):25-48.
［15］CHEN F, WANG R. Clustering algorithm for the fundamental matrix estimation ［J］. Journal of Computer-Aided Design and Computer Graphics, 2005,17(10):2251-2256.(陈付幸,王润生.基础矩阵估计的聚类分析算法［J］.计算机辅助设计与图形学学报,2005,17(10):2251-2256.)
［16］LU S, LEI Y, KONG W, et al.Robust fundamental matrix estimation based on kernel fuzzy clustering ［J］. Journal of Jilin University: Engineering and Technology Edition, 2012,42(2):434-439.(鲁珊,雷英杰,孔韦韦,等.基于模糊核聚类的鲁棒性基础矩阵估计算法［J］.吉林大学学报:工学版,2012,42(2):434-439.)
［17］FANG L. Research on feature based 3D scene reconstruction techniques from images sequence ［D］. Wuhan: Huazhong University of Science and Technology, 2006.(方磊.基于特征的图像序列三维场景重建技术研究［D］.武汉:华中科技大学,2006.)

[1]	杨鼎康黄帅王顺利翟鹏李一丹张立华. 基于对抗生成网络和网络集成的面部表情识别方法EE-GAN[J]. 计算机应用, 0, (): 0-0.
[2]	秦庭威赵鹏程秦品乐曾建朝柴锐黄永琦. 基于残差注意力机制的点云配准算法[J]. 计算机应用, 0, (): 0-0.
[3]	裴仪瑶, 郭会明, 张丹普, 陈文博. 基于定位不确定性的鲁棒3D目标检测方法[J]. 计算机应用, 2021, 41(10): 2979-2984.
[4]	许学斌, 张佳达, 刘伟, 路龙宾, 赵雨晴. 融合空间和通道特征的高精度乳腺癌分类方法[J]. 计算机应用, 2021, 41(10): 3025-3032.
[5]	汪虹余, 张彧, 杨恒, 穆楠. 基于蚁群优化算法的弱光图像显著性目标检测[J]. 计算机应用, 2021, 41(10): 2970-2978.
[6]	魏淳武, 赵涓涓, 唐笑先, 强彦. 基于多时期蒸馏网络的随访数据知识提取方法[J]. 计算机应用, 2021, 41(10): 2871-2878.
[7]	李自强, 王正勇, 陈洪刚, 李林怡, 何小海. 基于外观和动作特征双预测模型的视频异常行为检测[J]. 计算机应用, 2021, 41(10): 2997-3003.
[8]	张怡孙永荣赵科东李华曾庆化. 空中加油场景下的目标联合检测跟踪算法[J]. 计算机应用, 0, (): 0-0.
[9]	马峻姚震徐翠锋陈寿宏. 基于改进PP-YOLO和Deep-SORT的多无人机实时跟踪算法[J]. 计算机应用, 0, (): 0-0.
[10]	贺怀清闫建青惠康华. 基于深度残差网络的轻量级人脸识别方法[J]. 计算机应用, 0, (): 0-0.
[11]	王贺兵, 张春梅. 基于非对称卷积-压缩激发-次代残差网络的人脸关键点检测[J]. 计算机应用, 2021, 41(9): 2741-2747.
[12]	张笑非, 杨阳, 黄佳进, 钟宁. 基于度中心性的认知特征选择方法[J]. 计算机应用, 2021, 41(9): 2767-2772.
[13]	姬张建, 任兴旺. 带旋转与尺度估计的全卷积孪生网络目标跟踪算法[J]. 计算机应用, 2021, 41(9): 2705-2711.
[14]	闫钧华侯平张寅吕向阳马越王高飞. 基于多尺度多分类器卷积神经网络的混合失真类型判定方法 [J]. 计算机应用, 0, (): 0-0.
[15]	李福海蒋慕蓉杨磊谌俊毅. 基于生成对抗网络的梯度引导太阳斑点图像去模糊方法[J]. 计算机应用, 0, (): 0-0.

基于三视图约束的基础矩阵估计

Fundamental matrix estimation based on three-view constraint

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics