多视点深度图像的配准方法研究与实现
|
摘要
利用三维扫描技术来对物体三维重构是当今虚拟现实领域中的一项新技术,并且被广泛应用到案件现场三维重建、影视制作、文物保护等领域。为了重构三维物体的表面形状,必须事先得到物体表面的深度信息。但由于物体本身表面常常存在自遮挡和扫描角度误差等影响,只通过一次扫描不可能得到复杂物体表面的所有深度图像数据。必须通过扫描仪获得物体表面不同方向的深度图像;然后配准这些不同视角的深度图像,然后对其进行补洞、集成融合等操作,完成物体的三维建模。其中,深度图像配准是三维数字造型技术中的关键,它直接影响最后的合成结果、造型精度以及三维重构过程中的自动化程度。
本文首先简单介绍了利用三维扫描仪获取的深度图像对物体三维重构的一般流程,然后介绍了深度图像对应点集配准的算法原理,提出一种改进的配准算法。ICP(Iterative Closest Point)算法是目前最常用的深度图像配准算法,但ICP算法对两幅待配准深度图像初始相对位置要求较高,它们之间的初始位置不能相差太大,否则,ICP的收敛方向是不确定的,因而配准结果也是不可靠的,ICP算法很可能收敛于局部最优点,故初始相对位置对结果具有决定性影响,为此人们提出了一些基于ICP的优化算法。本文通过重点分析ICP以及其变种算法的优缺点,提出了一种“由粗到精”的配准思路,将遗传算法与改进后的ICP算法相结合运用到深度图像配准过程之中。初始配准主要利用遗传算法强大的全局最优搜索能力以及问题域的独立性和应用的鲁棒性特征,采用实数编码、通过实验确定了遗传算法对于配准问题的参数设置,缩小两幅深度图像之间的位置差,达到提高ICP算法稳定性的目的;在接下来的二次配准过程当中,从重叠区域检测、控制点选取、对应点集计算和对应点有效性检查等方面分别对原始的ICP算法提出了多种改进措施,实现了两幅深度图像之间快速精确的配准。实验表明,该算法易于实现、对待配准图像初始位置以及重叠部分的大小没有严格要求,有很好的鲁棒性,并且配准过程完全自动化、不需人为手动干预,提高了三维重构的效率。本文最后在基于上述算法的基础上,实现了案件现场三维重建与智能分析系统中三维重建部分子模块。
通过总结课题的研究工作,在本文最后还指明了下一步有必要深入开展的研究方向。
Three-dimensional objects reconstruction with 3D-scanning technology is a new technology in the field of Virtual reality, and this technology is widely used in 3D reconstruction of crime scenes, Television production, Heritage preservation and other fields. To reconstruct the three-dimensional object, one must get range images from the surface of the object. Confined by eye direction and the shape of an object, range images can not be acquired to describe the object by one scanning. To acquire a complete surface model of an object, one must scan the object from different views firstly, register range images of different views then and finally merge all range images in an unified coordinate. The key in the above process is the registration of range images, which influences the final result, the precision and automatation of 3D reconstruction.
The paper introduces the process of getting 3D models using range images obtained from the laser scanning simply, introduces the theory princeple of corresponding points registration and proposed an improved registration algorithm then. ICP(Iterative Closest Point) algorithem is the most widly used registration algorithem. Nevertheless, the proper convergence of ICP is guaranteed only if one of the datasets is a subset of the other; otherwise, erroneous alignments can result. Another drawback of ICP is that it requires a good pre-alignment of the views to converge to a correct solution. In order to solve these problems, researchers have proposed some improvements based on ICP. This paper focus on analyzing ICP algorithm, as well as the advantages and disadvantages of its variants,and proposed a "coarse-to-fine" registration thinking, combined Genetic Algorithm with improved ICP to the range registration. Rough registration mainly makes use of the Genetic Algorithm's powerful global optimal search capabilities, the independence of the problem domain and the robust nature of its application. Using real-coded ,we can determine the GAs parameter settings based on experiments for range registration. After rough registration, narrowed the location between two range images, improved the stability of ICP algorithm that will be implemented. In the next registration process, the paper introduces a fast iterative pairwise registration method, which combines four acceleration techniques: fast detection of overlapped regions, more-careful selection of control points, fast surface closet point computation and compatibility test of pairing points based on the difference of modulation. Experiments show that the new algorithm is easy to implement, with quadratic precision, it is not strict for initial position and the size of overlapping part, and robust. The process of range registration is fully automated, without human intervention, and can increase the efficiency of 3D reconstruction. Finally, based on the above algorithms, implemented the 3D reconstruction module in reconstruction and intelligence analysis system of crime scenes.
After summing up the research work, this paper points out some future research works.
本文首先简单介绍了利用三维扫描仪获取的深度图像对物体三维重构的一般流程,然后介绍了深度图像对应点集配准的算法原理,提出一种改进的配准算法。ICP(Iterative Closest Point)算法是目前最常用的深度图像配准算法,但ICP算法对两幅待配准深度图像初始相对位置要求较高,它们之间的初始位置不能相差太大,否则,ICP的收敛方向是不确定的,因而配准结果也是不可靠的,ICP算法很可能收敛于局部最优点,故初始相对位置对结果具有决定性影响,为此人们提出了一些基于ICP的优化算法。本文通过重点分析ICP以及其变种算法的优缺点,提出了一种“由粗到精”的配准思路,将遗传算法与改进后的ICP算法相结合运用到深度图像配准过程之中。初始配准主要利用遗传算法强大的全局最优搜索能力以及问题域的独立性和应用的鲁棒性特征,采用实数编码、通过实验确定了遗传算法对于配准问题的参数设置,缩小两幅深度图像之间的位置差,达到提高ICP算法稳定性的目的;在接下来的二次配准过程当中,从重叠区域检测、控制点选取、对应点集计算和对应点有效性检查等方面分别对原始的ICP算法提出了多种改进措施,实现了两幅深度图像之间快速精确的配准。实验表明,该算法易于实现、对待配准图像初始位置以及重叠部分的大小没有严格要求,有很好的鲁棒性,并且配准过程完全自动化、不需人为手动干预,提高了三维重构的效率。本文最后在基于上述算法的基础上,实现了案件现场三维重建与智能分析系统中三维重建部分子模块。
通过总结课题的研究工作,在本文最后还指明了下一步有必要深入开展的研究方向。
Three-dimensional objects reconstruction with 3D-scanning technology is a new technology in the field of Virtual reality, and this technology is widely used in 3D reconstruction of crime scenes, Television production, Heritage preservation and other fields. To reconstruct the three-dimensional object, one must get range images from the surface of the object. Confined by eye direction and the shape of an object, range images can not be acquired to describe the object by one scanning. To acquire a complete surface model of an object, one must scan the object from different views firstly, register range images of different views then and finally merge all range images in an unified coordinate. The key in the above process is the registration of range images, which influences the final result, the precision and automatation of 3D reconstruction.
The paper introduces the process of getting 3D models using range images obtained from the laser scanning simply, introduces the theory princeple of corresponding points registration and proposed an improved registration algorithm then. ICP(Iterative Closest Point) algorithem is the most widly used registration algorithem. Nevertheless, the proper convergence of ICP is guaranteed only if one of the datasets is a subset of the other; otherwise, erroneous alignments can result. Another drawback of ICP is that it requires a good pre-alignment of the views to converge to a correct solution. In order to solve these problems, researchers have proposed some improvements based on ICP. This paper focus on analyzing ICP algorithm, as well as the advantages and disadvantages of its variants,and proposed a "coarse-to-fine" registration thinking, combined Genetic Algorithm with improved ICP to the range registration. Rough registration mainly makes use of the Genetic Algorithm's powerful global optimal search capabilities, the independence of the problem domain and the robust nature of its application. Using real-coded ,we can determine the GAs parameter settings based on experiments for range registration. After rough registration, narrowed the location between two range images, improved the stability of ICP algorithm that will be implemented. In the next registration process, the paper introduces a fast iterative pairwise registration method, which combines four acceleration techniques: fast detection of overlapped regions, more-careful selection of control points, fast surface closet point computation and compatibility test of pairing points based on the difference of modulation. Experiments show that the new algorithm is easy to implement, with quadratic precision, it is not strict for initial position and the size of overlapping part, and robust. The process of range registration is fully automated, without human intervention, and can increase the efficiency of 3D reconstruction. Finally, based on the above algorithms, implemented the 3D reconstruction module in reconstruction and intelligence analysis system of crime scenes.
After summing up the research work, this paper points out some future research works.
引文
[1]潘云鹤.CAD系统与方法.杭州:浙江大学出版社,1996.
[2]孙家广.计算机辅助几何造型技术.北京:清华大学出版社,1990.
[3]唐荣锡.CAD/CAM技术.北京:北京航空航天大学出版社,1994
[4]殷国富等.计算机辅助设计技术与应用.北京:科学出版社,2000.
[5]Burdca G,Coiffet P.Virtual Reality Technology.A Wiley-interscience Publication,1994
[6]Carolina Cruz-Neirn.Vitual reality overview.Overview of virtual reality systems.Course Notes 23 of SIGGRAPH'93,1993
[7]Levoy,K.Pulli,B.Curless,S.Rusinkiewicz,et al.The Digital Ms.Mechelangelo Project:3D Scanning of Large Statues.In Proceedings of Siggraph 2000:131-144.
[8]Zhang Zonghua,Peng Xiang,Shi Weiqiang,et al.A survey of surface reconstruction from multiple range images[A].In:2nd Asia-Pacific Conference on Systems Integrity and Maintenance[C].Nanjing,2000:519-523.
[9]Besl P,McKay N.A Method for Registration of 3D Shapes.IEEE Trans on Pattern Analysis and Machine Intelligence,Vol.14,No.2,pp.239-256,Feberuary,1992
[10]Chen Y,Medioni G.Object modeling by registration of multiple range images[J].Image and Vision Computing,1992,10(3):145-155.
[11]G.Turk,M.Levoy.Zippered Polygon Meshes from Range Images.In Proceedings of SIGGRAPH,1994:311-318
[12]Chen.C.Z.Compact Surface Representation.Ph.D Dissertation.School of Information Technology and Engineering,University of Ottawa
[13]G.Dalley,P.Flynn.Pair-Wise Range Image Registration:A Study in Outlier Classification.Computer Vision and Image Understanding.2002,87(1):104-115
[14]N.Gelfand,L.Ikemoto,S.Rusinldewicz,et al.Geometrically stable sampling for the ICP algorithm.Proceedings of International Conference on 3D Digital Imaging and Modeling,Banff,2003,National Research Council of Canada,2003:260-267
[15]J.Jost,H.Hugli.A multi-resolution ICP with heuristic closest point search for fast and robust 3d registration of range images.Proceedings of International Conference on 3D Digital Imaging and Modeling,Banff,2003,National Research Council of Canada,2003:427-433
[16]N.J.Mitra,N.Gelfand,H.Pottmann,et al.Registration of point cloud data from a geometric optimization perspective.R.Scopigno,D.Zorin,Eurographics Symposium on Geometry Processing,Grenoble,2004.European Association for Computer Graphics,2004:22-31
[17]何文峰.大型场景三维重建中的深度图像配准:[工学硕士学位论文].北京:北京大学,2004
[18]章毓晋.图像工程(下册)图像理解.北京:清华大学出版社,2006
[19]J.A.Beraldin,F.Blais,L.Cournoyer,et al.Active 3D sensing,The Way into the Four Dimensions of Cultural Heritage Congress.Vienna,Austria.April 8-12,2003.
[20]E Blais.A Review of 20 Years of Range Sensor Development Videometrics Ⅶ,Proceedings of SPIE,Volume 5013,January 2003:62-76.
[21]Chert F,Brown GM and Song M Overview of three-dimensional shape measurement using optical methods Opt.2000:10-22.
[22]Boehler,W,A.Marbs.3D Scanning Instruments,CIPA,Heritage documentation International Workshop on Scanning for Cultural Heritage Recording Corfu,Greece.2002
[23]惠增宏.激光三维扫描、重建技术及其在工程中的应用:[工学硕士学位论文].西安:西北工业大学,2002
[24]陶曼.深度图像的分割与压缩:[工学硕士学位论文].北京:首都师范大学,2006
[25]贾云得.机器视觉.北京:科学出版社,2002
[26]S.Blostein and T.Huang.Estimating 3-D motion from range data in Proceedings first Conference on Artificial Intelligence Applications,Denver,December 1984:246-250
[27]K.S.Arum T.S.Huang and S.D.Bostein,Least-squares fitting of two3-D point sets.IEEE Trans.Part.Anal.And Mach.Intell.Vol.9,No.9,698-700 Sept.1987.
[28]J.K.Aggarwal and M.J.Magee.Determining motion parameters using intensity guided range sensing.Pattern Recognition.1986(19):169-180
[29]O.D.Faugeras,Q.T.Luong,S.J.Maybank.Camera Self-Calibration:Theory and Experiments.European Conference on Computer Vision,Santa Margherita Ligure, 1992:321-334
[30]B.K.P.Horn.Closed-form Solution of Absolute Orientation using unit quatemion.Journal of the Optical Society of America.1987,4(4):629-642
[31]K.Kanatani.Statistical Bias of Conic Fitting and Renormalization.1EEE Trans.PAMI.1994,16(3):320-326.
[32]D.W.Eggert,A.Lorusso,et al.Estimating 3-D rigid body transformations:a comparison of four major algorithms.Machine Vison and Applications.1997,9(5-6):272-290.
[33]G.Godin,M.Rioux and R.Baribeau.Three-dimensional registration using range and intensity information.In Videometries DI.Proceedings of SPIE 2350,1994,Nov,279-290
[34]Dorai.C,Weng.J,and Jain.A.Registration and Integration of Multiple Object Views for 3D Model Constrution.Trans.PAMI,Vol.20,No.1,1998.
[35]周军.多视点距离图像对准和集成方法的研究:[工学硕士论文].北京:北京工业大学,2001
[36]Z.Y.Zhang.Iterative point matching for registration of free-form curves and surfaces.International Journal of Computer Vision.1994,13(2):119-152.
[37]E.Bittar,S.Lavallee and R.Szeliski.A Method for Registering overlapping range images of arbitrarily shaped surfaces for 3-D object reconstruction.In:Sensor Fusion Ⅵ.Proceedings of SPIE 059,1993:384-395.
[38]宋万忠.光学三维传感中多视点距离像配准与复杂曲面展平:[博士学位论文].成都:四川大学,2002
[39]Faugeras and M.Hebert,The representation,recognition and locating of 3D objects.Int.J.Robotics Res.S,1986:27-52
[40]R.J.Campbell and P.J.Flynn.A survey of free-form object representation and recognition techniques.Computer Vision and Image Understanding,81(2):166-210,2001.
[41]Pulli,K.Multiview Registration for Large Data Sets.Pros.3DIM,1999.
[42]Rodriguez.MA,Yonghuai Liu,Qian Wei.Geometric alignment of two overlapping range images Acoustics,Speech and Signal Processing,2001.Proceedings.(ICASSP'O1).2001IEEE International Conference on,Volume:3,7-11 May 2001:1689-1692
[43]Rodriguez.MA,Yonghuai Liu.Registering two overlapping range images using a relative registration error histogram Image Processing.2002 International Conference on,Volume: 3,24-28 June 2002:841-844
[44]李敏强,林丹,李书全.遗传算法的基本理论与应用.北京:科学出版社,2002.
[45]刘正东.计算机视觉中立体匹配技术的研究:[博士学位论文].南京:南京理工大学,2005
[46]Tom.P,and Zisserman.A.MLESAC:A new robust estimator with application to estimating image geometry,Computer Vision and Image Understanding,78,1,2000:138-156.
[47]Gotardo.P,Bellon.O,and Silva.L.Range image segmentation by surface extraction using an improved robust estimator.Proceedings of IEEE Conference on Corrzputer Vision and Pattern Recognition,2,2003:33-38.
[48]Gotardo.P,Bellon.O,Silva.L,et al.Range image segmentation by surface extraction using improved robust estimator and genetic algorithm,IEEE Transactions on Sgsterrzs,Man and bernetics,Part B.Submitted.2003
[49]Cinque.L,Corzani.F,Levialdi.S,et al.Improvement in range segmentation parameters tuning.International Conference on Patern Recognition,1,2002:176-179.
[50]Man.K,Tang.K,and Kwong.S.Genetic algorithms:concepts and applications,IEEE Zhansactions on Industrial Eletronics,43,5,1996:519-534.
[51]Haralick R M,Jee H,Lee CN,et al.Pose Lstimation from Corresponding Point Data,IEEE trans.PAMI,1989:1426-1446
[52]P.Rousseeuw,S.Van Aelst,Positive-Breakdown Robust Methods in ComputerVision,Computing Science and Statistics 31,1999:451-460.
[53]P.Meer,D.Mintz,A.Rosenfedl,D.Kim,Robust regression methods in computer vision:a review,International Journal of Computer Vision 6,1991:59-70.
[54]M.Ye,R.Haralick,Optical Flow From A Least-Trimmed Squares Based Adaptive Approach,in:Proc.International Conf.on Pattern Recognition,Vol.3,IEEE Compter Soc.,2000:1064-1067.
[55]D.Chetverikov,Fast neighborhood search in planar point set,Pattern Recognition Letters 12,1991:409-412.
[56]W.Press,S.Teukolsky,W.Vetterling,et al.Numerical Recipes in C,Cambridge University Press,1992.
[2]孙家广.计算机辅助几何造型技术.北京:清华大学出版社,1990.
[3]唐荣锡.CAD/CAM技术.北京:北京航空航天大学出版社,1994
[4]殷国富等.计算机辅助设计技术与应用.北京:科学出版社,2000.
[5]Burdca G,Coiffet P.Virtual Reality Technology.A Wiley-interscience Publication,1994
[6]Carolina Cruz-Neirn.Vitual reality overview.Overview of virtual reality systems.Course Notes 23 of SIGGRAPH'93,1993
[7]Levoy,K.Pulli,B.Curless,S.Rusinkiewicz,et al.The Digital Ms.Mechelangelo Project:3D Scanning of Large Statues.In Proceedings of Siggraph 2000:131-144.
[8]Zhang Zonghua,Peng Xiang,Shi Weiqiang,et al.A survey of surface reconstruction from multiple range images[A].In:2nd Asia-Pacific Conference on Systems Integrity and Maintenance[C].Nanjing,2000:519-523.
[9]Besl P,McKay N.A Method for Registration of 3D Shapes.IEEE Trans on Pattern Analysis and Machine Intelligence,Vol.14,No.2,pp.239-256,Feberuary,1992
[10]Chen Y,Medioni G.Object modeling by registration of multiple range images[J].Image and Vision Computing,1992,10(3):145-155.
[11]G.Turk,M.Levoy.Zippered Polygon Meshes from Range Images.In Proceedings of SIGGRAPH,1994:311-318
[12]Chen.C.Z.Compact Surface Representation.Ph.D Dissertation.School of Information Technology and Engineering,University of Ottawa
[13]G.Dalley,P.Flynn.Pair-Wise Range Image Registration:A Study in Outlier Classification.Computer Vision and Image Understanding.2002,87(1):104-115
[14]N.Gelfand,L.Ikemoto,S.Rusinldewicz,et al.Geometrically stable sampling for the ICP algorithm.Proceedings of International Conference on 3D Digital Imaging and Modeling,Banff,2003,National Research Council of Canada,2003:260-267
[15]J.Jost,H.Hugli.A multi-resolution ICP with heuristic closest point search for fast and robust 3d registration of range images.Proceedings of International Conference on 3D Digital Imaging and Modeling,Banff,2003,National Research Council of Canada,2003:427-433
[16]N.J.Mitra,N.Gelfand,H.Pottmann,et al.Registration of point cloud data from a geometric optimization perspective.R.Scopigno,D.Zorin,Eurographics Symposium on Geometry Processing,Grenoble,2004.European Association for Computer Graphics,2004:22-31
[17]何文峰.大型场景三维重建中的深度图像配准:[工学硕士学位论文].北京:北京大学,2004
[18]章毓晋.图像工程(下册)图像理解.北京:清华大学出版社,2006
[19]J.A.Beraldin,F.Blais,L.Cournoyer,et al.Active 3D sensing,The Way into the Four Dimensions of Cultural Heritage Congress.Vienna,Austria.April 8-12,2003.
[20]E Blais.A Review of 20 Years of Range Sensor Development Videometrics Ⅶ,Proceedings of SPIE,Volume 5013,January 2003:62-76.
[21]Chert F,Brown GM and Song M Overview of three-dimensional shape measurement using optical methods Opt.2000:10-22.
[22]Boehler,W,A.Marbs.3D Scanning Instruments,CIPA,Heritage documentation International Workshop on Scanning for Cultural Heritage Recording Corfu,Greece.2002
[23]惠增宏.激光三维扫描、重建技术及其在工程中的应用:[工学硕士学位论文].西安:西北工业大学,2002
[24]陶曼.深度图像的分割与压缩:[工学硕士学位论文].北京:首都师范大学,2006
[25]贾云得.机器视觉.北京:科学出版社,2002
[26]S.Blostein and T.Huang.Estimating 3-D motion from range data in Proceedings first Conference on Artificial Intelligence Applications,Denver,December 1984:246-250
[27]K.S.Arum T.S.Huang and S.D.Bostein,Least-squares fitting of two3-D point sets.IEEE Trans.Part.Anal.And Mach.Intell.Vol.9,No.9,698-700 Sept.1987.
[28]J.K.Aggarwal and M.J.Magee.Determining motion parameters using intensity guided range sensing.Pattern Recognition.1986(19):169-180
[29]O.D.Faugeras,Q.T.Luong,S.J.Maybank.Camera Self-Calibration:Theory and Experiments.European Conference on Computer Vision,Santa Margherita Ligure, 1992:321-334
[30]B.K.P.Horn.Closed-form Solution of Absolute Orientation using unit quatemion.Journal of the Optical Society of America.1987,4(4):629-642
[31]K.Kanatani.Statistical Bias of Conic Fitting and Renormalization.1EEE Trans.PAMI.1994,16(3):320-326.
[32]D.W.Eggert,A.Lorusso,et al.Estimating 3-D rigid body transformations:a comparison of four major algorithms.Machine Vison and Applications.1997,9(5-6):272-290.
[33]G.Godin,M.Rioux and R.Baribeau.Three-dimensional registration using range and intensity information.In Videometries DI.Proceedings of SPIE 2350,1994,Nov,279-290
[34]Dorai.C,Weng.J,and Jain.A.Registration and Integration of Multiple Object Views for 3D Model Constrution.Trans.PAMI,Vol.20,No.1,1998.
[35]周军.多视点距离图像对准和集成方法的研究:[工学硕士论文].北京:北京工业大学,2001
[36]Z.Y.Zhang.Iterative point matching for registration of free-form curves and surfaces.International Journal of Computer Vision.1994,13(2):119-152.
[37]E.Bittar,S.Lavallee and R.Szeliski.A Method for Registering overlapping range images of arbitrarily shaped surfaces for 3-D object reconstruction.In:Sensor Fusion Ⅵ.Proceedings of SPIE 059,1993:384-395.
[38]宋万忠.光学三维传感中多视点距离像配准与复杂曲面展平:[博士学位论文].成都:四川大学,2002
[39]Faugeras and M.Hebert,The representation,recognition and locating of 3D objects.Int.J.Robotics Res.S,1986:27-52
[40]R.J.Campbell and P.J.Flynn.A survey of free-form object representation and recognition techniques.Computer Vision and Image Understanding,81(2):166-210,2001.
[41]Pulli,K.Multiview Registration for Large Data Sets.Pros.3DIM,1999.
[42]Rodriguez.MA,Yonghuai Liu,Qian Wei.Geometric alignment of two overlapping range images Acoustics,Speech and Signal Processing,2001.Proceedings.(ICASSP'O1).2001IEEE International Conference on,Volume:3,7-11 May 2001:1689-1692
[43]Rodriguez.MA,Yonghuai Liu.Registering two overlapping range images using a relative registration error histogram Image Processing.2002 International Conference on,Volume: 3,24-28 June 2002:841-844
[44]李敏强,林丹,李书全.遗传算法的基本理论与应用.北京:科学出版社,2002.
[45]刘正东.计算机视觉中立体匹配技术的研究:[博士学位论文].南京:南京理工大学,2005
[46]Tom.P,and Zisserman.A.MLESAC:A new robust estimator with application to estimating image geometry,Computer Vision and Image Understanding,78,1,2000:138-156.
[47]Gotardo.P,Bellon.O,and Silva.L.Range image segmentation by surface extraction using an improved robust estimator.Proceedings of IEEE Conference on Corrzputer Vision and Pattern Recognition,2,2003:33-38.
[48]Gotardo.P,Bellon.O,Silva.L,et al.Range image segmentation by surface extraction using improved robust estimator and genetic algorithm,IEEE Transactions on Sgsterrzs,Man and bernetics,Part B.Submitted.2003
[49]Cinque.L,Corzani.F,Levialdi.S,et al.Improvement in range segmentation parameters tuning.International Conference on Patern Recognition,1,2002:176-179.
[50]Man.K,Tang.K,and Kwong.S.Genetic algorithms:concepts and applications,IEEE Zhansactions on Industrial Eletronics,43,5,1996:519-534.
[51]Haralick R M,Jee H,Lee CN,et al.Pose Lstimation from Corresponding Point Data,IEEE trans.PAMI,1989:1426-1446
[52]P.Rousseeuw,S.Van Aelst,Positive-Breakdown Robust Methods in ComputerVision,Computing Science and Statistics 31,1999:451-460.
[53]P.Meer,D.Mintz,A.Rosenfedl,D.Kim,Robust regression methods in computer vision:a review,International Journal of Computer Vision 6,1991:59-70.
[54]M.Ye,R.Haralick,Optical Flow From A Least-Trimmed Squares Based Adaptive Approach,in:Proc.International Conf.on Pattern Recognition,Vol.3,IEEE Compter Soc.,2000:1064-1067.
[55]D.Chetverikov,Fast neighborhood search in planar point set,Pattern Recognition Letters 12,1991:409-412.
[56]W.Press,S.Teukolsky,W.Vetterling,et al.Numerical Recipes in C,Cambridge University Press,1992.