基于图像的数字城市自动化重建技术研究
|
摘要
三维数字城市是一项跨学科的系统工程。尽管在理论研究方面已取得一些进展，但在普通计算机上实现实时显示和交互仍有很大难度。探索低成本、高效率的城市三维模型自动重建途径，对信息技术发展尤其是智能城市建设具有重要的理论意义和实践价值。
针对开发数字城市系统存在的挑战，本文从图像重建的角度出发，讨论了建筑物重建、DEM数据重建、建筑物定位等问题，取得的研究成果有：
1．基于单幅照片的建筑物重建研究。主要研究了单个建筑物的重建方法，一是提出一种改进的小波ICA滤波器，运用ICA规范小波降维后的信号，发现独立噪声的特征，从图像源中分离噪声与信号。二是基于GA-PSO Hough变换提出一种建筑物平面的重构方法，采用有效的限制搜索空间方法以及分层次搜索策略，有效降低了计算量，提高了重构性能与质量。三是提出从3D建筑物图像中纹理提取与映射的方法，通过校正输入图像，融合透视图提取纹理图片；并提出消除透视扭曲的方法，对纹理图像进行校正，通过补偿光源颜色以估计纹理色彩，避免了3D测量的中间过程。
2．基于立体图片的城市重建研究。主要研究了地形与建筑物的重建方法，一是根据线阵推扫式影像的几何性质，分析了线阵推扫式影像的核线模型，提出基于核线曲线的共轭匹配方法，解决了共轭搜索算法、相关域的确定和前后相关匹配策略等问题，并通过实验总结分析了算法的优越性。二是利用自动模糊聚类法进行建筑物抽取，运用Hopfield神经立体匹配方法，对经典匹配进行初始化，提高了匹配率并减小了不明确性。三是针对立体图像匹配时任务调度效率低的缺点，提出了一种启发式任务调度算法HCDDSL，优化任务调度的结果，减少任务完成时间。
3．基于特征的模型识别与定位研究。主要研究了大场景图片中建筑物的识别与定位方法，一是提出应用小波变换进行输入图像的边提取方法，提取目标的边缘特征。利用多层次小波性质，实现不同种类目标的检测。二是基于几何不变关系，提出一种三维目标识别算法，通过摄像机投影变换，分析投影参数约束，使用两种不变关系，进行视场不变的目标识别。三是应用联合估计模型和同步联合估计模型，提出了一种基于能量的联合运动和视差估计算法，利用左右运动和前一帧的视差来计算联合估计约束当前帧的初始视差。
3D digital city is a systematic engineering concerning with multi subjects. Even thoughsome progresses have been made in theoretical research, it is still very difficult with regard toreal-time display and interactive manipulation on general computers. Therefore, for thedevelopment of information technology and the construction of intelligent city, it ismeaningful both theoretically and practically to explore the three dimensional automaticreconstruction method of cities.
As far as the challenges in developing digital city is concerned, from imagereconstruction point of view, building reconstruction, DEM data reconstruction and buildinglocation are elaborated in this thesis. The work and results are expressed as follows.
1. Reconstruct the building with single frame photography. The main research is thereconstruction method of sigle building. Firstly, an improved wavelet-ICA filter, which canextract the independent noise feature, is proposed. ICA is employed to regularize the signalfrom the wavelet dimensionality reduction of original image and serves to identify the noisefeature. Secondly, GA-PSO is incorporated to improve the Hough transform (HT)performance. Different strategies are introduced to reduce the computational effort, whichinclude efficient limitation of the search space and a hierarchical search method. A modifiedGA-PSO HT algorithm is developed to improve the performance and quality of reconstructionfor planar building recognition techniques. Thirdly, a method dealing with texture extractionand mapping form3D building image is also proposed, in which the3D-building isreconstructed based on the planar layout. Texture images are then extracted by mergingperspective and correcting input images. A new technique is studied and presented for thepurpose of removing perspective distortion and estimating the texture image color tocompensate for the light source color. The method is free of any intermediate process in3Dmeasurement.
2. Reconstruct the city with stereo photograph. The main research is the reconstructionmethod of terrain and buildings. On the first place, according to the geometric properties oflinear pushbroom images, the epipolarity model of Linear CCD pushbroom images isanalyzed. A conjugate matching method is proposed on the basis of epipolar curve. The conjugate search algorithm, the correlation region solution and match strategy are solved withthe conjugate matching method. The advantage of this method is validated with experiments.Then features of the building are extracted with proposed Fuzzy threshold method. Hopfieldneural stereo matching method is set up and used to initialize classical matching problems.The proposed Hopfield neural stereo matching method improves matching rate and decreaseambiguities simultaneously. On the third place, the shortcomings of existing task schedulingalgorithms are analyzed, and a new efficient heuristic task scheduling algorithm, namely,HCDDSL is proposed. Through HCDDSL, the result of task schedule may be optimizedeffectively, and the time span of all tasks is also reduced.
3. The model recognition and locating are accomplished with the features. The mainresearch is the identification and location method for building in large scene picture. At first,the wavelet transform method is utilized to process the edge extraction. The edge feature ofthe target is extracted through proposed method. Different kinds of targets detection can berealized with multi-level wavelet properties. Secondly, a new3D target recognition algorithmis proposed based on geometrically invariant relationships. The algorithm employs a cameramodel for projective image formation and the projection parameters constraints are analyzed.Besides, two invariant relationships are utilized to complete the target recognition withview-invariant object. Finally, an energy-based estimation algorithm uniting motion and viewdisparity is proposed on the basis of two energy models, which are the joint estimation modeland the simultaneous joint estimation model. The initial disparity of the current frame iscomputed with the difference between side-to-side motion and previous frame.
针对开发数字城市系统存在的挑战，本文从图像重建的角度出发，讨论了建筑物重建、DEM数据重建、建筑物定位等问题，取得的研究成果有：
1．基于单幅照片的建筑物重建研究。主要研究了单个建筑物的重建方法，一是提出一种改进的小波ICA滤波器，运用ICA规范小波降维后的信号，发现独立噪声的特征，从图像源中分离噪声与信号。二是基于GA-PSO Hough变换提出一种建筑物平面的重构方法，采用有效的限制搜索空间方法以及分层次搜索策略，有效降低了计算量，提高了重构性能与质量。三是提出从3D建筑物图像中纹理提取与映射的方法，通过校正输入图像，融合透视图提取纹理图片；并提出消除透视扭曲的方法，对纹理图像进行校正，通过补偿光源颜色以估计纹理色彩，避免了3D测量的中间过程。
2．基于立体图片的城市重建研究。主要研究了地形与建筑物的重建方法，一是根据线阵推扫式影像的几何性质，分析了线阵推扫式影像的核线模型，提出基于核线曲线的共轭匹配方法，解决了共轭搜索算法、相关域的确定和前后相关匹配策略等问题，并通过实验总结分析了算法的优越性。二是利用自动模糊聚类法进行建筑物抽取，运用Hopfield神经立体匹配方法，对经典匹配进行初始化，提高了匹配率并减小了不明确性。三是针对立体图像匹配时任务调度效率低的缺点，提出了一种启发式任务调度算法HCDDSL，优化任务调度的结果，减少任务完成时间。
3．基于特征的模型识别与定位研究。主要研究了大场景图片中建筑物的识别与定位方法，一是提出应用小波变换进行输入图像的边提取方法，提取目标的边缘特征。利用多层次小波性质，实现不同种类目标的检测。二是基于几何不变关系，提出一种三维目标识别算法，通过摄像机投影变换，分析投影参数约束，使用两种不变关系，进行视场不变的目标识别。三是应用联合估计模型和同步联合估计模型，提出了一种基于能量的联合运动和视差估计算法，利用左右运动和前一帧的视差来计算联合估计约束当前帧的初始视差。
3D digital city is a systematic engineering concerning with multi subjects. Even thoughsome progresses have been made in theoretical research, it is still very difficult with regard toreal-time display and interactive manipulation on general computers. Therefore, for thedevelopment of information technology and the construction of intelligent city, it ismeaningful both theoretically and practically to explore the three dimensional automaticreconstruction method of cities.
As far as the challenges in developing digital city is concerned, from imagereconstruction point of view, building reconstruction, DEM data reconstruction and buildinglocation are elaborated in this thesis. The work and results are expressed as follows.
1. Reconstruct the building with single frame photography. The main research is thereconstruction method of sigle building. Firstly, an improved wavelet-ICA filter, which canextract the independent noise feature, is proposed. ICA is employed to regularize the signalfrom the wavelet dimensionality reduction of original image and serves to identify the noisefeature. Secondly, GA-PSO is incorporated to improve the Hough transform (HT)performance. Different strategies are introduced to reduce the computational effort, whichinclude efficient limitation of the search space and a hierarchical search method. A modifiedGA-PSO HT algorithm is developed to improve the performance and quality of reconstructionfor planar building recognition techniques. Thirdly, a method dealing with texture extractionand mapping form3D building image is also proposed, in which the3D-building isreconstructed based on the planar layout. Texture images are then extracted by mergingperspective and correcting input images. A new technique is studied and presented for thepurpose of removing perspective distortion and estimating the texture image color tocompensate for the light source color. The method is free of any intermediate process in3Dmeasurement.
2. Reconstruct the city with stereo photograph. The main research is the reconstructionmethod of terrain and buildings. On the first place, according to the geometric properties oflinear pushbroom images, the epipolarity model of Linear CCD pushbroom images isanalyzed. A conjugate matching method is proposed on the basis of epipolar curve. The conjugate search algorithm, the correlation region solution and match strategy are solved withthe conjugate matching method. The advantage of this method is validated with experiments.Then features of the building are extracted with proposed Fuzzy threshold method. Hopfieldneural stereo matching method is set up and used to initialize classical matching problems.The proposed Hopfield neural stereo matching method improves matching rate and decreaseambiguities simultaneously. On the third place, the shortcomings of existing task schedulingalgorithms are analyzed, and a new efficient heuristic task scheduling algorithm, namely,HCDDSL is proposed. Through HCDDSL, the result of task schedule may be optimizedeffectively, and the time span of all tasks is also reduced.
3. The model recognition and locating are accomplished with the features. The mainresearch is the identification and location method for building in large scene picture. At first,the wavelet transform method is utilized to process the edge extraction. The edge feature ofthe target is extracted through proposed method. Different kinds of targets detection can berealized with multi-level wavelet properties. Secondly, a new3D target recognition algorithmis proposed based on geometrically invariant relationships. The algorithm employs a cameramodel for projective image formation and the projection parameters constraints are analyzed.Besides, two invariant relationships are utilized to complete the target recognition withview-invariant object. Finally, an energy-based estimation algorithm uniting motion and viewdisparity is proposed on the basis of two energy models, which are the joint estimation modeland the simultaneous joint estimation model. The initial disparity of the current frame iscomputed with the difference between side-to-side motion and previous frame.
引文
[1] L.Alvarez,P.L.Lions,J.M.Morel.Image selective smoothing and edge detection bynonlinear diffusion[A].SIAM1992,1992(29)：845–866P
[2] J.F.Aujol,G.Aubert,L.Blanc-F′eraud,A.Chambolle.Image decomposition intoa bounded variation component and an oscillating component[J].Math. Imaging Vision,2005(22)：71–88P
[3] S.A.Awate,R.T.Whitaker.Image denoising with unsupervised information-theoretic adaptive filtering[A].the IEEE International Conference on ComputerVision and Pattern Recognition，2005
[4] A.Buades,B.Coll,J.M.Morel.Image data process by image noise reduction andcamera integrating the means for implementingthis process[A].French patent application.
[5] F.Catt′e,F.Dibos,G.Koepfler.A morphological scheme for mean curvature motio-n and applications to anisotropic diffusion and motion of level sets[A].SIAM1995，1995(32)：1875–1909P
[6] A.Chambolle,P.L.Lions.Image recovery via total variation minimization and rela-ted problems[J].Numer Math，1997(76)：167–188P
[7] T.F.Chan,H.M.Zhou.Total variation improved wavelet thresholding in imagecompression[A].the IEEE International Conference on Image Processing，2000（2）：391–394P
[8]刘传才,杨静宇.一种新的图像纹理表示方法[J].计算机学报,2001,14(11)，1202-1209页
[9] R.R.Coifman,D.Donoho.Translation-invariant de-noising in Wavelets and Statisti-cs[M].Springer-Verlag,1995：125–150P
[10] D. Donoho.Denoising by soft-thresholding[A].IEEE Transform Theory,1995(41)：613–627P
[11] S.Loncric.A survey of shape analysis techniques[A].Pattern Recognition,1998(8)：983-1001P
[12]顶险峰，吴洪，张宏江.形状匹配综述[J].自动化学报,2001,27(9)：678-694页
[13] D.Donoho,I.Johnstone.Ideal spatial adaptation via wavelet shrinkage[A].Biometrika,1994(81)：425–455P
[14] S.Durand,M.Nikolova.Restoration of wavelet coefficients by minimizing a specially designed objective function[A].the IEEE Workshop on Variational and Level Set Methods in Computer Vision,2003.
[15] S.Durand,J.Froment.Reconstruction of wavelet coefficients using total variationminimization[J].SIAM J.Sci.Comput2003,2003(24)：1754–1767P
[16] A.Efros,T.Leung.Texture synthesis by non parametric sampling[A].Proceedingsof the IEEE International Conference on Computer Vision1999,1999(2)：1033–1038P
[17] L.G. Brown.A survey of image registration techniques[J].ACM Comput.Surv,1992,24,(4)：325–376P
[18]熊惠霖，张天序.具有平移和尺度不变性的自适应小波变换[J].电子学报，2000(1)：104–107页
[19] S.L.Tanimoto.Template matching in pyramids[J].Comput.Graphics Image Process.1981(16)：356–369P
[20]杨翔英，章毓晋.小波轮廓描述符及在图像查询中的应用[J].计算机学报，1999(7)：752-757页
[21] S.-H. Lai，M. Fang.Robust and efficient image alignment with spatially varyingillumination model[A].Proceedings of the Computer Vision and Pattern Recognition1999，1999(7)：167–172P
[22] C.F.Olson，D.P.Huttenlocher.Automatic target recognition by matching oriented edge pixels[A].IEEE Trans. Image Process，1997：103–113P
[23] D.H. Ballard.Generalizing the Hough transform to detect arbitrary shapes[J].Pattern Recognition.1981，13(2)：111–122P
[24]姚玉荣，章毓晋.利用小波和矩进行基于形状的图像检索[J].中国图像图形学报，2000(3)：206-210页
[25] L.S. Davis.Hierarchical generalized Hough transforms and line-segment based generalized Hough transforms[J].Pattern Recognition，1982，15(4)：277–285P.
[26]吴四九.图像目标形状匹配与视频滤波技术研究[J].成都信息工程学院学报，2007,17(2)
[27] V. Cantoni，L. Carrioli.Structural shape recognition in a multiresolution environment[J].Signal Process，1987(12)：267–276P
[28] W. Lu，J.L. Tan.Detection of incomplete ellipse in images with strong noise byiterative randomized Hough transform[J].Pattern Recognition，2008，41(4)：1268–1279P
[29]李飚，李吉成.一种基于神经网络的红外目标识别方法[J].系统工程与电子技术，1999,21(2)
[30] A. Torii，A. Imiya.The randomized-Hough-transform-based method for great-circle detection on sphere[J].Pattern Recognition Letters，2007,28(10)：1186–1192P.
[31]陈振羽,李德华,周焰.利用多种特征和Hopfield神经网络的有遮挡的目标识别[J].中国图像图形学报，2000，5(12)：1034-1038P
[32] E.Duquenoy，A.Taleb-Ahmed.Applying the Hough transform pseudo-linearity property to improve computing speed[J].Pattern Recognition Letters，2006,7(16)：1893–1904P
[33] C.Baillard，A.Zisserman.Automatic reconstruction of piecewise planar models from multiple views[A].Proceedings of the Conference on Computer Vision and Pattern Recognition，1999：559–565P
[34] P.E.Debevec, C.J.Taylor, J.Malik.Modeling and rendering architecture from photographs: a hybrid geometry-and image-based approach[A].Siggraph，1996
[35]刘怡光，游志胜.一种用于图像目标识别的神经网络及其车型识别应用[J].计算机工程，2003,29(3)：30-32页
[36] A.R. Dick, P.H.S. Torr, S.F. Ruffle, R. Cipolla.Combining single view recognition and multiple view stereo for architectural scenes[A].Proceedings of the8th International Conference on Computer Vision，2001.
[37] A.W. Fitzgibbon, G. Cross, A. Zisserman.Automatic3D model construction forturn-table sequences[A].Proceedings of the SMILE Workshop on3D Structurefrom Multiple Images of Large-Scale Environments，1998.
[38]孔海东，于明，韩立法.基于边缘与灰度梯度的人眼特征定位分析[J].河北工业大学学报，2008,37(3).
[39] P. Gargallo, P. Sturm.Bayesian3D modeling from images using multiple depthmaps[A].Proceedings of the International Conference on Computer Vision and Pattern Recognition，2005.
[40] C.Hernandez Esteban, F.Schmitt.Multi-stereo3D object reconstruction[A].International Symposium on3D Data Processing Visualization and Transmission，2002：159–166P
[41] C.Hernandez Esteban, F.Schmitt.Silhouette and stereo fusion for3D object modeling[A].Fourth International Conference on3D Digital Imaging and Modeling，2003:46–53P
[42]李向吉，丁润涛，蔡靖.边缘检测器性能的客观定量评价[J].电子科学学刊，1999，21(5).
[43] A.Fitzgibbon，G.Cross，A.Zisserman.Automatic3D model construction for turn-table sequences.European Workshop SMILE'98，1998：155–170P
[44]郭军，倪晋平，王会峰等.一种使用小波矩的飞行器目标识别方法[J].西安工业大学学报，2008，28(2).
[45] T. Fromherz，M. Bichsel.Multiple depth and normal maps for shape from multiple views and visual cues[A].ISPRS Intercommission Workshop From Pixels toSequences，1995
[46] M.Kampel，S.Tosovic，R.Sablatnig.Octree-based fusion of shape from silhouetteand shape from structured light[A].First IEEE International Symposium on3DData Processing Visualization and Transmission, Padova，2002：754–757P
[47]李学勇，路长厚，李国平.基于二阶梯度图的Canny检测边缘修补方法[J].光电子·激光，2007,18(3)
[48] Anitha Panapakkam，S.N. Balakrishnan.Proceeding of SPIE.995：10P
[49] Wang Mian，Wang Jingjing，Wang Bo，Wang Wensheng.Acta Armamentar ofChina，2006(27)：836P
[50]罗钧，侯艳，付丽.一种改进的灰度矩亚像素边缘检测算法[J].重庆大学学报：自然科学版，2008,31(5)
[51] Renu Tripathi，Kehar Singh.Optical Engineering[C]：532P
[52] Alpana Bhagatji, Naveen K. Nishchal, Arun K. Gupta，B.P. Tyagi.Optics Communications[C]，2008：44P
[53]张桂敏，蒋晓瑜，郭广生等.一种基于图像灰度值切线方向的边缘检测方法[J].装甲兵工程学院学报，2007,21(2)
[54] Wensheng Wang，Yu. Chen，Chuiping Liang，Hua Miao.Proceeding of SPIE5642，2004：204P
[55] Mei Xie，Zheng Ma.Acta Electronic Sinic，1999(27)：106P
[56] Xu Qiong.Journal of Sichuan Institute of Light Industry and Chemical Technology，1998：11P
[57]肖家庆，卢凌，李晟等.基于矩不变量的图像目标识别[J].武汉理工大学学报：交通科学与工程版，2006,30(4)
[58] Guo Lianxi，Guo Xian jiu.Journal of Liaoning Normal University (Natural Science Edition)，2006(29)：181P
[59] Guorong Gao, Ran Liu and Xuming Yi.Journal of Wuhan University: Natural Science Edition，2005(51)：615P
[60] Barrett，E.B.Payton，P.M.Haag, N.N. Brill.General methods for determining projective invariants in imagery[C].CVGIP: Image Understanding，1991：46–65P
[61]胡学龙，程茜，杨莉.基于形态滤波和梯度模糊CHMT的组合边缘检测[J]电子测量与仪器学报，2008,22(3)
[62]武妍，魏广英，葛菊萍等.基于形态学双梯度算子的边缘检测[J].计算机工程与应用，2007,43(36)
[63] Burns, J.B,Weiss, R.S,Riseman, E.M.View variation of point-set and line-segment features[C].IEEE Trans Pattern Anal，1993：51–68P
[64]胡硕，朱明，吴川.结合Zernike矩的多尺度模板形状匹配[J].光电工程，2005，32(10)
[65] Carlsson,S.Projectively invariant decomposition and recognition of planar shapes[C]. Int.J. Comput. Vision，1996：193–209P
[66] Forsyth,D.A.Mundy, J.L.Zisserman, A.P.Coelho, C.Heller,Aand Rothwell, C.A. Invariant descriptors for3D object recognition and pose[C]. IEEE Trans PatternAnal，1991：971–991P
[67]崔屹.图像处理与分析－数学形态学方法及应用[M]，北京：科学出版社.2005
[68] Haralick, R.M.Using perspective transform in scene analysis[J]. ComputGraphicsImage Process，1981：191–221P
[69] Hartley.R，Zisserman.A.Multiple View Geometry in Computer Vision[M].Cambridge University Press, Cambridge, UK，2000
[70]丁德锋.运动目标的检测、识别与跟踪技术研究[D].西北工业大学硕士论文，2007：20-50页
[71] Lutton, E.Maitre, H.Lopez-Krahe.Contribution to the determination of vanishingpoints using Hough transform[A].IEEE Trans Pattern Analy，1994：430–483P
[72] Long Quan.Invariants of six points and projective reconstruction from three uncalibrated images[A].IEEE Trans Pattern Anal，1995：34–46P
[73] Rothwell，C.A. Forsyth，D.A. Zisserman, A. Mundy.Extracting projective structure from single perspective views of3D point sets.Proceedings of IEEE International Conference on Computer Vision，1993：573–582P
[74]王耀明，严炜，董建萍.图像的亮度矩和阈值分割[J].上海师范大学学报(自然科学版)，2001，30(1)
[75] Weiss.Noise-resistant invariants of curves[A]. IEEE Trans. Pattern Anal，1993：943–948P
[76] Weiss.Geometric invariants and object recognition[A]. Int. J. Comput，1993：207–231.
[77] Weiss, Ray. Model-based recognition of3D objects from one view[A].Proceedings of Fifth European Conference on Computer Vision，1998：716–732P
[78] Huang J，Zhang M，Ma J.Spectral quadrangulation with orientation and alignment control[J].ACM Transactions on Graphics，2008，27(5)：1-9P
[79] Zhu Y，Seneviratne..New algorithm for calculating an invariant of3D point sets from a single view[J].Image Vision Comput，1996：179–188P
[80] Branch JW, Prieto F, Boulanger P.Automatic extraction of quadrilateral patchesfrom triangulated surface using Morse theory[C].In Proceedings of the16thInternational Meshing Roundtable，2008：190-212P
[81] Kos G，Martin RR,，Varady T.Methods to recover constant radius rolling ballblends in reverse engineering[J].Computer Aided Geometric Design，2000,17(2)：127-160P
[82]王飞，万少松，袁华.红外/毫米波雷达双模导引末状态的目标跟踪算法[J].兵工自动化，2006，25(10)
[83]张丽莉，姚勇，刘志镜.一种相对势能分水岭算法[J].西安电子科技大学学报，2008,35(1)
[84] Zhang L，Zhou R，Zhu J. Piecewise B-spline surfaces fitting to arbitrary triangle meshes[J].CIRP Annals-Manufacturing Technology，2002，51(1)：131-134P
[85] Pauly M.Point Primitives for Interactive Modeling and Processing of3D Geometry[D].Federal Institute of Technology of Zurich，2003.
[86] Flynn PJ，Jain AK.On reliable curvature estimation[C].In Proceedings of IEEEConference on Computer Vision and Pattern Recognition，1989：110-116P
[87] Eck M，Hoppe H.Automatic reconstruction of B-spline surfaces of arbitrary topological type[C].In Proceedings of the23rdAnnual Conference on Computer Graphics and Interactive Techniques，1996:325-333P
[88] Bauer U，Polthier K.Generating parametric models of tubes from laser scans[J].Computer-Aided Design，2009，41(10)：719-729P
[89] Nieser M，Schulz C，Polthier K.Patch layout from feature graphs[J].Computer-Aided Design，2010,42(3)：213-220P
[90] Taubin G.Estimating the tensor of curvature of a surface from a polyhedral approximation[C].In Proceedings of the Fifth International Conference on Computer Vision，1995：902-907P
[91] Chen X，Schmitt F.Intrinsic surface properties from surface triangulation[C].InProceedings of European Conference on Computer Vision，1992：739-743P
[92] Hameiri E，Shimshoni I.Estimating the principal curvatures and the darboux frame from real3D range data[C].In First International Symposium on3D DataProcessing Visualization and Transmission，2002：626-637P
[93] Meyer N, Mathieu D, Schroder P, et al.Discrete differential-geometry operatorsfor triangulated2-manifolds[C]. In Visualization and Mathematics，2002:34-57P
[94] Gray A.Modern Differential Geometry of Curves and Surfaces with Mathematics[M]. Boca Raton, FL: CRC Press，1998.
[95] Miao YW, Feng JQ, Peng QS. Curvature estimation of point-sampled surfacesand its applications[J]. ICCSA2005, Lecture Notes in Computer Science，2005,3482:1023-1032P
[96] Ernest MS, Shang YW.Surface Parameterization and Curvature Measurement ofArbitrary3-D Objects: Five Practical Method [J]. IEEE Transactions on PatternAnalysis and Machine Intelligence，1992,14(8):833-839P
[97]柯映林，陈曦.基于4D Shepard曲面的点云曲率估算[J].浙江大学学报，2005,39(6)：761-764页
[98] Yang M，Lee E.Segmentation of measured point data using a parametric quadric surface approximation [J].Computer-Aided Design，1999,31(7):449-457P
[99]贺美芳，周来水，神会存.散乱点云数据的曲率估算及应用[J].南京航空航天大学学报，2005,37(4):515-519页
[100]吴剑煌，刘伟军，王天然等.点采样曲面的曲率估计[J].仪器仪表学报，2006,27(12)：1557-1562页
[101] Razdan A, Bae M.Curvature estimation scheme for triangle meshes using biquadratic Bézier patches[J].Computer-Aided Design，2005，37(14)：1481-1491P
[102]孙殿柱，范志先，李延瑞等.散乱数据点云型面特征分析算法的研究与应用[J].机械工程学报，2007,43(6)：133-136页
[103] Leonardis A，Gupta A，Bajscy R.Segmentation of range images as the search for geometric parametric models[J].International Journal of Computer Vision，1995,14(3)：253-277P
[104] Yang M, Lee E.Segmentation of measured point data using a parametric quadric surface approximation[J].Computer-Aided Design，1999,31(7)：449-457P
[105] Barhak J，Fischer A.Parameterization for reconstruction of3D freeform objectsfrom laser-scanned data based on a PDE method[J].Visual Computer，2001，17(6)：353-369P
[106]史桂蓉，邢渊，李永清.反向工程应用现状及研究方向[J].机械科学与技术，2000,19(4)：653-654页
[107] Huang J，Menq C.Automatic data segmentation for geometric feature extractionfrom unorganized3-D coordinate points[J].IEEE Transactions on Robotics andAutomation，2001，17(3)：268-279P
[108] Ke Y，Fan S，Zhu W，et al.Feature-based reverse modeling strategies[J].Computer-Aided Design，2006,38(5).485-506P
[109] Shah J.Assessment of features technology[J].Computer-Aided Design，1991,23(5)：331-343P
[110]金涛，单岩，童水光.产品反求工程中基于几何特征及约束的模型重建[J].计算机辅助设计与图形学学报，2001,13(3)：202-207页
[111]刘胜兰.逆向工程中自由曲面与规则曲面重建关键技术研究[D].南京航空航天大学博士学位论文，2004.
[112]柯映林,李岸.点云数据中拉伸面特征的提取[J].计算机辅助设计与图形学学报，2005,17(6)：1329-1334页
[113] Matthew D，Ronald S.Determining the axis of a surface of revolution using tactile sensing[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，1989,15(10)：1079-1087P
[114] Werghi N，Fisher R，Robertson C，et al.Object reconstruction by incorporatinggeometricconstraints in reverse engineering[J].Computer-Aided Design，1999,31(6)：363-399P
[115] Benko P，Martin RR，Varady T.Algorithms for reverse engineering boundary representation models[J].Computer-Aided Design，2001,33(11)：839-851P
[116] Pottmann H，Randrup T.Rotational and helical surface approximation for reverseengineering[J].Computing，1998，60(4)：307-322P
[117]孙玉文，贾振元.基于自由曲面点云的快速原型制作技术研究[J].机械工程学报，2003，39(1)：56-59页
[118] Pottmann H，Peternell M，Ravani B.An introduction to line geometry with application[J].Computer-Aided Design，1999,31(2)：3-16P
[119] Lai JY，Ueng WD.Reconstruction of surfaces of revolution from measured points[J].Computers in Industry，2000,41(2).147-161P
[120] Patane G，Falcidieno B.Defining, contouring, and visualizing scalar functions onpoint-sampled surfaces[J].Computer-Aided Design，2011,43(3)：227-246P
[2] J.F.Aujol,G.Aubert,L.Blanc-F′eraud,A.Chambolle.Image decomposition intoa bounded variation component and an oscillating component[J].Math. Imaging Vision,2005(22)：71–88P
[3] S.A.Awate,R.T.Whitaker.Image denoising with unsupervised information-theoretic adaptive filtering[A].the IEEE International Conference on ComputerVision and Pattern Recognition，2005
[4] A.Buades,B.Coll,J.M.Morel.Image data process by image noise reduction andcamera integrating the means for implementingthis process[A].French patent application.
[5] F.Catt′e,F.Dibos,G.Koepfler.A morphological scheme for mean curvature motio-n and applications to anisotropic diffusion and motion of level sets[A].SIAM1995，1995(32)：1875–1909P
[6] A.Chambolle,P.L.Lions.Image recovery via total variation minimization and rela-ted problems[J].Numer Math，1997(76)：167–188P
[7] T.F.Chan,H.M.Zhou.Total variation improved wavelet thresholding in imagecompression[A].the IEEE International Conference on Image Processing，2000（2）：391–394P
[8]刘传才,杨静宇.一种新的图像纹理表示方法[J].计算机学报,2001,14(11)，1202-1209页
[9] R.R.Coifman,D.Donoho.Translation-invariant de-noising in Wavelets and Statisti-cs[M].Springer-Verlag,1995：125–150P
[10] D. Donoho.Denoising by soft-thresholding[A].IEEE Transform Theory,1995(41)：613–627P
[11] S.Loncric.A survey of shape analysis techniques[A].Pattern Recognition,1998(8)：983-1001P
[12]顶险峰，吴洪，张宏江.形状匹配综述[J].自动化学报,2001,27(9)：678-694页
[13] D.Donoho,I.Johnstone.Ideal spatial adaptation via wavelet shrinkage[A].Biometrika,1994(81)：425–455P
[14] S.Durand,M.Nikolova.Restoration of wavelet coefficients by minimizing a specially designed objective function[A].the IEEE Workshop on Variational and Level Set Methods in Computer Vision,2003.
[15] S.Durand,J.Froment.Reconstruction of wavelet coefficients using total variationminimization[J].SIAM J.Sci.Comput2003,2003(24)：1754–1767P
[16] A.Efros,T.Leung.Texture synthesis by non parametric sampling[A].Proceedingsof the IEEE International Conference on Computer Vision1999,1999(2)：1033–1038P
[17] L.G. Brown.A survey of image registration techniques[J].ACM Comput.Surv,1992,24,(4)：325–376P
[18]熊惠霖，张天序.具有平移和尺度不变性的自适应小波变换[J].电子学报，2000(1)：104–107页
[19] S.L.Tanimoto.Template matching in pyramids[J].Comput.Graphics Image Process.1981(16)：356–369P
[20]杨翔英，章毓晋.小波轮廓描述符及在图像查询中的应用[J].计算机学报，1999(7)：752-757页
[21] S.-H. Lai，M. Fang.Robust and efficient image alignment with spatially varyingillumination model[A].Proceedings of the Computer Vision and Pattern Recognition1999，1999(7)：167–172P
[22] C.F.Olson，D.P.Huttenlocher.Automatic target recognition by matching oriented edge pixels[A].IEEE Trans. Image Process，1997：103–113P
[23] D.H. Ballard.Generalizing the Hough transform to detect arbitrary shapes[J].Pattern Recognition.1981，13(2)：111–122P
[24]姚玉荣，章毓晋.利用小波和矩进行基于形状的图像检索[J].中国图像图形学报，2000(3)：206-210页
[25] L.S. Davis.Hierarchical generalized Hough transforms and line-segment based generalized Hough transforms[J].Pattern Recognition，1982，15(4)：277–285P.
[26]吴四九.图像目标形状匹配与视频滤波技术研究[J].成都信息工程学院学报，2007,17(2)
[27] V. Cantoni，L. Carrioli.Structural shape recognition in a multiresolution environment[J].Signal Process，1987(12)：267–276P
[28] W. Lu，J.L. Tan.Detection of incomplete ellipse in images with strong noise byiterative randomized Hough transform[J].Pattern Recognition，2008，41(4)：1268–1279P
[29]李飚，李吉成.一种基于神经网络的红外目标识别方法[J].系统工程与电子技术，1999,21(2)
[30] A. Torii，A. Imiya.The randomized-Hough-transform-based method for great-circle detection on sphere[J].Pattern Recognition Letters，2007,28(10)：1186–1192P.
[31]陈振羽,李德华,周焰.利用多种特征和Hopfield神经网络的有遮挡的目标识别[J].中国图像图形学报，2000，5(12)：1034-1038P
[32] E.Duquenoy，A.Taleb-Ahmed.Applying the Hough transform pseudo-linearity property to improve computing speed[J].Pattern Recognition Letters，2006,7(16)：1893–1904P
[33] C.Baillard，A.Zisserman.Automatic reconstruction of piecewise planar models from multiple views[A].Proceedings of the Conference on Computer Vision and Pattern Recognition，1999：559–565P
[34] P.E.Debevec, C.J.Taylor, J.Malik.Modeling and rendering architecture from photographs: a hybrid geometry-and image-based approach[A].Siggraph，1996
[35]刘怡光，游志胜.一种用于图像目标识别的神经网络及其车型识别应用[J].计算机工程，2003,29(3)：30-32页
[36] A.R. Dick, P.H.S. Torr, S.F. Ruffle, R. Cipolla.Combining single view recognition and multiple view stereo for architectural scenes[A].Proceedings of the8th International Conference on Computer Vision，2001.
[37] A.W. Fitzgibbon, G. Cross, A. Zisserman.Automatic3D model construction forturn-table sequences[A].Proceedings of the SMILE Workshop on3D Structurefrom Multiple Images of Large-Scale Environments，1998.
[38]孔海东，于明，韩立法.基于边缘与灰度梯度的人眼特征定位分析[J].河北工业大学学报，2008,37(3).
[39] P. Gargallo, P. Sturm.Bayesian3D modeling from images using multiple depthmaps[A].Proceedings of the International Conference on Computer Vision and Pattern Recognition，2005.
[40] C.Hernandez Esteban, F.Schmitt.Multi-stereo3D object reconstruction[A].International Symposium on3D Data Processing Visualization and Transmission，2002：159–166P
[41] C.Hernandez Esteban, F.Schmitt.Silhouette and stereo fusion for3D object modeling[A].Fourth International Conference on3D Digital Imaging and Modeling，2003:46–53P
[42]李向吉，丁润涛，蔡靖.边缘检测器性能的客观定量评价[J].电子科学学刊，1999，21(5).
[43] A.Fitzgibbon，G.Cross，A.Zisserman.Automatic3D model construction for turn-table sequences.European Workshop SMILE'98，1998：155–170P
[44]郭军，倪晋平，王会峰等.一种使用小波矩的飞行器目标识别方法[J].西安工业大学学报，2008，28(2).
[45] T. Fromherz，M. Bichsel.Multiple depth and normal maps for shape from multiple views and visual cues[A].ISPRS Intercommission Workshop From Pixels toSequences，1995
[46] M.Kampel，S.Tosovic，R.Sablatnig.Octree-based fusion of shape from silhouetteand shape from structured light[A].First IEEE International Symposium on3DData Processing Visualization and Transmission, Padova，2002：754–757P
[47]李学勇，路长厚，李国平.基于二阶梯度图的Canny检测边缘修补方法[J].光电子·激光，2007,18(3)
[48] Anitha Panapakkam，S.N. Balakrishnan.Proceeding of SPIE.995：10P
[49] Wang Mian，Wang Jingjing，Wang Bo，Wang Wensheng.Acta Armamentar ofChina，2006(27)：836P
[50]罗钧，侯艳，付丽.一种改进的灰度矩亚像素边缘检测算法[J].重庆大学学报：自然科学版，2008,31(5)
[51] Renu Tripathi，Kehar Singh.Optical Engineering[C]：532P
[52] Alpana Bhagatji, Naveen K. Nishchal, Arun K. Gupta，B.P. Tyagi.Optics Communications[C]，2008：44P
[53]张桂敏，蒋晓瑜，郭广生等.一种基于图像灰度值切线方向的边缘检测方法[J].装甲兵工程学院学报，2007,21(2)
[54] Wensheng Wang，Yu. Chen，Chuiping Liang，Hua Miao.Proceeding of SPIE5642，2004：204P
[55] Mei Xie，Zheng Ma.Acta Electronic Sinic，1999(27)：106P
[56] Xu Qiong.Journal of Sichuan Institute of Light Industry and Chemical Technology，1998：11P
[57]肖家庆，卢凌，李晟等.基于矩不变量的图像目标识别[J].武汉理工大学学报：交通科学与工程版，2006,30(4)
[58] Guo Lianxi，Guo Xian jiu.Journal of Liaoning Normal University (Natural Science Edition)，2006(29)：181P
[59] Guorong Gao, Ran Liu and Xuming Yi.Journal of Wuhan University: Natural Science Edition，2005(51)：615P
[60] Barrett，E.B.Payton，P.M.Haag, N.N. Brill.General methods for determining projective invariants in imagery[C].CVGIP: Image Understanding，1991：46–65P
[61]胡学龙，程茜，杨莉.基于形态滤波和梯度模糊CHMT的组合边缘检测[J]电子测量与仪器学报，2008,22(3)
[62]武妍，魏广英，葛菊萍等.基于形态学双梯度算子的边缘检测[J].计算机工程与应用，2007,43(36)
[63] Burns, J.B,Weiss, R.S,Riseman, E.M.View variation of point-set and line-segment features[C].IEEE Trans Pattern Anal，1993：51–68P
[64]胡硕，朱明，吴川.结合Zernike矩的多尺度模板形状匹配[J].光电工程，2005，32(10)
[65] Carlsson,S.Projectively invariant decomposition and recognition of planar shapes[C]. Int.J. Comput. Vision，1996：193–209P
[66] Forsyth,D.A.Mundy, J.L.Zisserman, A.P.Coelho, C.Heller,Aand Rothwell, C.A. Invariant descriptors for3D object recognition and pose[C]. IEEE Trans PatternAnal，1991：971–991P
[67]崔屹.图像处理与分析－数学形态学方法及应用[M]，北京：科学出版社.2005
[68] Haralick, R.M.Using perspective transform in scene analysis[J]. ComputGraphicsImage Process，1981：191–221P
[69] Hartley.R，Zisserman.A.Multiple View Geometry in Computer Vision[M].Cambridge University Press, Cambridge, UK，2000
[70]丁德锋.运动目标的检测、识别与跟踪技术研究[D].西北工业大学硕士论文，2007：20-50页
[71] Lutton, E.Maitre, H.Lopez-Krahe.Contribution to the determination of vanishingpoints using Hough transform[A].IEEE Trans Pattern Analy，1994：430–483P
[72] Long Quan.Invariants of six points and projective reconstruction from three uncalibrated images[A].IEEE Trans Pattern Anal，1995：34–46P
[73] Rothwell，C.A. Forsyth，D.A. Zisserman, A. Mundy.Extracting projective structure from single perspective views of3D point sets.Proceedings of IEEE International Conference on Computer Vision，1993：573–582P
[74]王耀明，严炜，董建萍.图像的亮度矩和阈值分割[J].上海师范大学学报(自然科学版)，2001，30(1)
[75] Weiss.Noise-resistant invariants of curves[A]. IEEE Trans. Pattern Anal，1993：943–948P
[76] Weiss.Geometric invariants and object recognition[A]. Int. J. Comput，1993：207–231.
[77] Weiss, Ray. Model-based recognition of3D objects from one view[A].Proceedings of Fifth European Conference on Computer Vision，1998：716–732P
[78] Huang J，Zhang M，Ma J.Spectral quadrangulation with orientation and alignment control[J].ACM Transactions on Graphics，2008，27(5)：1-9P
[79] Zhu Y，Seneviratne..New algorithm for calculating an invariant of3D point sets from a single view[J].Image Vision Comput，1996：179–188P
[80] Branch JW, Prieto F, Boulanger P.Automatic extraction of quadrilateral patchesfrom triangulated surface using Morse theory[C].In Proceedings of the16thInternational Meshing Roundtable，2008：190-212P
[81] Kos G，Martin RR,，Varady T.Methods to recover constant radius rolling ballblends in reverse engineering[J].Computer Aided Geometric Design，2000,17(2)：127-160P
[82]王飞，万少松，袁华.红外/毫米波雷达双模导引末状态的目标跟踪算法[J].兵工自动化，2006，25(10)
[83]张丽莉，姚勇，刘志镜.一种相对势能分水岭算法[J].西安电子科技大学学报，2008,35(1)
[84] Zhang L，Zhou R，Zhu J. Piecewise B-spline surfaces fitting to arbitrary triangle meshes[J].CIRP Annals-Manufacturing Technology，2002，51(1)：131-134P
[85] Pauly M.Point Primitives for Interactive Modeling and Processing of3D Geometry[D].Federal Institute of Technology of Zurich，2003.
[86] Flynn PJ，Jain AK.On reliable curvature estimation[C].In Proceedings of IEEEConference on Computer Vision and Pattern Recognition，1989：110-116P
[87] Eck M，Hoppe H.Automatic reconstruction of B-spline surfaces of arbitrary topological type[C].In Proceedings of the23rdAnnual Conference on Computer Graphics and Interactive Techniques，1996:325-333P
[88] Bauer U，Polthier K.Generating parametric models of tubes from laser scans[J].Computer-Aided Design，2009，41(10)：719-729P
[89] Nieser M，Schulz C，Polthier K.Patch layout from feature graphs[J].Computer-Aided Design，2010,42(3)：213-220P
[90] Taubin G.Estimating the tensor of curvature of a surface from a polyhedral approximation[C].In Proceedings of the Fifth International Conference on Computer Vision，1995：902-907P
[91] Chen X，Schmitt F.Intrinsic surface properties from surface triangulation[C].InProceedings of European Conference on Computer Vision，1992：739-743P
[92] Hameiri E，Shimshoni I.Estimating the principal curvatures and the darboux frame from real3D range data[C].In First International Symposium on3D DataProcessing Visualization and Transmission，2002：626-637P
[93] Meyer N, Mathieu D, Schroder P, et al.Discrete differential-geometry operatorsfor triangulated2-manifolds[C]. In Visualization and Mathematics，2002:34-57P
[94] Gray A.Modern Differential Geometry of Curves and Surfaces with Mathematics[M]. Boca Raton, FL: CRC Press，1998.
[95] Miao YW, Feng JQ, Peng QS. Curvature estimation of point-sampled surfacesand its applications[J]. ICCSA2005, Lecture Notes in Computer Science，2005,3482:1023-1032P
[96] Ernest MS, Shang YW.Surface Parameterization and Curvature Measurement ofArbitrary3-D Objects: Five Practical Method [J]. IEEE Transactions on PatternAnalysis and Machine Intelligence，1992,14(8):833-839P
[97]柯映林，陈曦.基于4D Shepard曲面的点云曲率估算[J].浙江大学学报，2005,39(6)：761-764页
[98] Yang M，Lee E.Segmentation of measured point data using a parametric quadric surface approximation [J].Computer-Aided Design，1999,31(7):449-457P
[99]贺美芳，周来水，神会存.散乱点云数据的曲率估算及应用[J].南京航空航天大学学报，2005,37(4):515-519页
[100]吴剑煌，刘伟军，王天然等.点采样曲面的曲率估计[J].仪器仪表学报，2006,27(12)：1557-1562页
[101] Razdan A, Bae M.Curvature estimation scheme for triangle meshes using biquadratic Bézier patches[J].Computer-Aided Design，2005，37(14)：1481-1491P
[102]孙殿柱，范志先，李延瑞等.散乱数据点云型面特征分析算法的研究与应用[J].机械工程学报，2007,43(6)：133-136页
[103] Leonardis A，Gupta A，Bajscy R.Segmentation of range images as the search for geometric parametric models[J].International Journal of Computer Vision，1995,14(3)：253-277P
[104] Yang M, Lee E.Segmentation of measured point data using a parametric quadric surface approximation[J].Computer-Aided Design，1999,31(7)：449-457P
[105] Barhak J，Fischer A.Parameterization for reconstruction of3D freeform objectsfrom laser-scanned data based on a PDE method[J].Visual Computer，2001，17(6)：353-369P
[106]史桂蓉，邢渊，李永清.反向工程应用现状及研究方向[J].机械科学与技术，2000,19(4)：653-654页
[107] Huang J，Menq C.Automatic data segmentation for geometric feature extractionfrom unorganized3-D coordinate points[J].IEEE Transactions on Robotics andAutomation，2001，17(3)：268-279P
[108] Ke Y，Fan S，Zhu W，et al.Feature-based reverse modeling strategies[J].Computer-Aided Design，2006,38(5).485-506P
[109] Shah J.Assessment of features technology[J].Computer-Aided Design，1991,23(5)：331-343P
[110]金涛，单岩，童水光.产品反求工程中基于几何特征及约束的模型重建[J].计算机辅助设计与图形学学报，2001,13(3)：202-207页
[111]刘胜兰.逆向工程中自由曲面与规则曲面重建关键技术研究[D].南京航空航天大学博士学位论文，2004.
[112]柯映林,李岸.点云数据中拉伸面特征的提取[J].计算机辅助设计与图形学学报，2005,17(6)：1329-1334页
[113] Matthew D，Ronald S.Determining the axis of a surface of revolution using tactile sensing[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，1989,15(10)：1079-1087P
[114] Werghi N，Fisher R，Robertson C，et al.Object reconstruction by incorporatinggeometricconstraints in reverse engineering[J].Computer-Aided Design，1999,31(6)：363-399P
[115] Benko P，Martin RR，Varady T.Algorithms for reverse engineering boundary representation models[J].Computer-Aided Design，2001,33(11)：839-851P
[116] Pottmann H，Randrup T.Rotational and helical surface approximation for reverseengineering[J].Computing，1998，60(4)：307-322P
[117]孙玉文，贾振元.基于自由曲面点云的快速原型制作技术研究[J].机械工程学报，2003，39(1)：56-59页
[118] Pottmann H，Peternell M，Ravani B.An introduction to line geometry with application[J].Computer-Aided Design，1999,31(2)：3-16P
[119] Lai JY，Ueng WD.Reconstruction of surfaces of revolution from measured points[J].Computers in Industry，2000,41(2).147-161P
[120] Patane G，Falcidieno B.Defining, contouring, and visualizing scalar functions onpoint-sampled surfaces[J].Computer-Aided Design，2011,43(3)：227-246P