基于关键点特征描述子的三维文物碎片重组
|
摘要
对多个文物碎片进行重组时,确定文物碎片之间的邻接关系时间复杂度非常高;文物断裂部位因受损而存在一定的几何特征缺失,仅依靠断裂区域的几何特征进行拼接或重组难免会出现错误.鉴于此,提出一种基于关键点特征描述子的三维文物碎片重组方法.首先利用已有的文物"模板",采用内在形状签名算法提取文物"模板"和文物碎片原始面上的关键点;然后计算关键点的快速点特征直方图特征描述子;再通过比较其相似性来确定文物碎片在"模板"上的对应位置,即确定各个碎片断裂面之间的邻接关系;最后采用顺序拼合的方式及迭代最近点算法对多个碎片进行重组.实验结果表明,该方法能够有效地提高文物碎片虚拟拼接的效率.
The time complexity was very high for determining adjacency relationship of the fragments, when reassembling archaeological fragments. Due to damaged, there was somewhat geometrical characteristic missing in the fracture surface of cultural relics. It can lead to some errors for matching or reassembling. This paper proposed an effective reassembly pipeline based on feature descriptors of key points. Firstly, the integrated template with similar geometry to the patchwork was available in some reassembly tasks, and Intrinsic Shape Signature(ISS) algorithm was used to extract key points on intact regions of the model and fractured fragment respectively. Then, the Fast Point Feature Histograms(FPFH) feature descriptors were calculated; The adjacency relations between the fracture surfaces were determined by comparing the similarity of FPFH feature descriptor. Finally, multiple fragments were reorganized by sequential splicing and Iterative Closest Point(ICP)algorithm was used to achieve fine registration. The experimental results show that the proposed method can improve efficiency of 3 D archaeological fragment reassembly.
The time complexity was very high for determining adjacency relationship of the fragments, when reassembling archaeological fragments. Due to damaged, there was somewhat geometrical characteristic missing in the fracture surface of cultural relics. It can lead to some errors for matching or reassembling. This paper proposed an effective reassembly pipeline based on feature descriptors of key points. Firstly, the integrated template with similar geometry to the patchwork was available in some reassembly tasks, and Intrinsic Shape Signature(ISS) algorithm was used to extract key points on intact regions of the model and fractured fragment respectively. Then, the Fast Point Feature Histograms(FPFH) feature descriptors were calculated; The adjacency relations between the fracture surfaces were determined by comparing the similarity of FPFH feature descriptor. Finally, multiple fragments were reorganized by sequential splicing and Iterative Closest Point(ICP)algorithm was used to achieve fine registration. The experimental results show that the proposed method can improve efficiency of 3 D archaeological fragment reassembly.
引文
[1]Cooper D B, Willis A, Andrews S, et al. Bayesian pot-assembly fromfragmentsasproblemsinperceptual-groupingandgeometric-learning[C]//Proceedingsofthe16thInternational Conference on Pattern Recognition. Los Alamitos:IEEE Computer Society Press, 2002, 3:297-302
[2]McBride J C, Kimia B B. Archaeological fragment reconstruction using curve-matching[C]//Proceedings of Conference on ComputerVisionandPatternRecognitionWorkshop.Los Alamitos:IEEE Computer Society Press, 2003, 1:3
[3]Papaioannou G, Karabassi E A. On the automatic assemblage of arbitrary broken solid artefacts[J]. Image and Vision Computing, 2003, 21(5):401-412
[4]Huang Q X, Fl?ry S, Gelfand N, et al. Reassembling fractured objectsbygeometricmatching[J].ACMTransactionson Graphics, 2006, 25(3):569-578
[5]Zhou Shucheng, Geng Guohua, Zhou Mingquan. A multiscale methodformosaicing3Dfragmentedobjects[J].Journalof Computer-Aided Design&Computer Graphics, 2006, 18(10):1525-1530(in Chinese)(周术诚,耿国华,周明全.三维破碎物体多尺度拼接技术[J].计算机辅助设计与图形学学报, 2006, 18(10):1525-1530)
[6]Li Chunlong, Zhou Mingquan, Cheng Xin, et al. Virtual restorationofaxisymmetricrelicfragments[J].JournalofComputer-AidedDesign&ComputerGraphics,2006,18(5):620-624(in Chinese)(李春龙,周明全,成欣,等.轴对称破碎文物的虚拟复原方法[J].计算机辅助设计与图形学学报, 2006, 18(5):620-624)
[7]Papaodysseus C, Arabadjis D, Exarhos M, et al. Efficient solutiontothe3Dproblemofautomaticwallpaintingsreassembly[J].Computers&MathematicswithApplications,2012,64(8):2712-2734
[8]Li Qunhui, Zhou Mingquan, Geng Guohua. Reassembly of broken 3D solids based on fractured surfaces matching[J]. Journal of Image and Graphics, 2012, 17(10):1298-1304(in Chinese)(李群辉,周明全,耿国华.断裂面匹配的破碎刚体复原[J].中国图象图形学报, 2012, 17(10):1298-1304)
[9]LiuJun,ZhouMingquan,GengGuohua,etal.Fragments splicing method for terra-cotta figures of Qin Dynasty based on contoursandfracturesurfacesmatching[J].ComputerEngineering, 2014, 40(1):181-185+190(in Chinese)(刘军,周明全,耿国华,等.基于轮廓与断面匹配的秦俑碎片拼接方法[J].计算机工程, 2014, 40(1):181-185+190)
[10]Li Shanshan, Geng Guohua, Zhou Mingquan, et al. Interactive reassembly of fractured fragments based on surface adjacency constraint[J]. Journal of Computer-Aided Design&Computer Graphics, 2016, 28(6):924-931(in Chinese)(李姗姗,耿国华,周明全,等.基于表面邻接约束的交互式文物碎片重组[J].计算机辅助设计与图形学学报,2016,28(6):924-931)
[11]Zhou Mingquan, Yuan Jie, Geng Guohua, et al. Interactive reassemblyoffracturedfragmentsbasedonfeaturepointsof contour line[J]. Optics and Precision Engineering, 2017, 25(6):1597-1606(in Chinese)(周明全,袁洁,耿国华,等.基于轮廓线特征点的交互式文物拼接[J].光学精密工程, 2017, 25(6):1597-1606)
[12]YuanJie,ZhouMingquan,GengGuohua,etal.Heritagedebris splicing algorithm based on contour line two-way distance field[J].Computer Engineering, 2018, 44(6):207-212+218(in Chinese)(袁洁,周明全,耿国华,等.基于轮廓线双向距离场的文物碎片拼接算法[J].计算机工程, 2018, 44(6):207-212+218)
[13]Zhong Y. Intrinsic shape signatures:a shape descriptor for 3D objectrecognition[C]//Proceedingsofthe12thInternational ConferenceonComputerVisionWorkshops.LosAlamitos:IEEE Computer Society Press, 2009:689-696
[14]Rusu R B, Blodow N, Beetz M. Fast point feature histograms(FPFH)for 3D registration[C]//Proceedings of the IEEE International Conference on Robotics and Automation. Los Alamitos:IEEE Computer Society Press, 2009:1848-1853
[15]Fischler M A, Bolles R C. Random sample consensus:a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM, 1981,24(6):381-395
[16]BeslPJ,MckayHD.Amethodforregistrationof3-D shapes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992, 14(2):239-256
[17]Johnson A E, Hebert M. Using spin images for efficient object recognition in cluttered 3D scenes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1999, 21(5):433-449
[18]BelongieS,MalikJ.Matchingwithshapecontexts[C]//Proceedings of the IEEE Workshop on Content-Based Access of Image and Video Libraries. Los Alamitos:IEEE Computer Society Press, 2000:20
[19]Sipiran I, Bustos B. Harris 3D:a robust extension of the Harris operator for interest point detection on 3D meshes[J]. The Visual Computer, 2011, 27(11):963-976
[20]Rusu R B, Marton Z C, Blodow N, et al. Learning informative pointclassesfortheacquisitionofobjectmodelmaps[C]//Proceedings of the 10th International Conference on Control,Automation, Robotics andVision. LosAlamitos:IEEE Computer Society Press, 2008:643-650
[21]TombariF,SaltiS,StefanoLD.Uniquesignaturesofhistograms for local surface description[C]//Proceedings of the 11th EuropeanConferenceonComputerVision.Heidelberg:Springer, 2010, Part III:356-369
[22]Chen Z, Czarnuch S, Smith A, et al. Performance evaluation of 3D keypointsand descriptors[C]//Proceedings ofInternationalSymposium on Visual Computing. Heidelberg:Springer, 2016:410-420
[23]Yuan Jie, Zhou Mingquan, Geng Guohua, et al. Automatic reassembly of fractured fragments using morse topological features[J].Acta Automatica Sinica, 2018, 44(8):1486-1495(in Chinese)(袁洁,周明全,耿国华,等.基于Morse-Smale拓扑特征的文物碎片拼接算法[J].自动化学报, 2018, 44(8):1486-1495)
[2]McBride J C, Kimia B B. Archaeological fragment reconstruction using curve-matching[C]//Proceedings of Conference on ComputerVisionandPatternRecognitionWorkshop.Los Alamitos:IEEE Computer Society Press, 2003, 1:3
[3]Papaioannou G, Karabassi E A. On the automatic assemblage of arbitrary broken solid artefacts[J]. Image and Vision Computing, 2003, 21(5):401-412
[4]Huang Q X, Fl?ry S, Gelfand N, et al. Reassembling fractured objectsbygeometricmatching[J].ACMTransactionson Graphics, 2006, 25(3):569-578
[5]Zhou Shucheng, Geng Guohua, Zhou Mingquan. A multiscale methodformosaicing3Dfragmentedobjects[J].Journalof Computer-Aided Design&Computer Graphics, 2006, 18(10):1525-1530(in Chinese)(周术诚,耿国华,周明全.三维破碎物体多尺度拼接技术[J].计算机辅助设计与图形学学报, 2006, 18(10):1525-1530)
[6]Li Chunlong, Zhou Mingquan, Cheng Xin, et al. Virtual restorationofaxisymmetricrelicfragments[J].JournalofComputer-AidedDesign&ComputerGraphics,2006,18(5):620-624(in Chinese)(李春龙,周明全,成欣,等.轴对称破碎文物的虚拟复原方法[J].计算机辅助设计与图形学学报, 2006, 18(5):620-624)
[7]Papaodysseus C, Arabadjis D, Exarhos M, et al. Efficient solutiontothe3Dproblemofautomaticwallpaintingsreassembly[J].Computers&MathematicswithApplications,2012,64(8):2712-2734
[8]Li Qunhui, Zhou Mingquan, Geng Guohua. Reassembly of broken 3D solids based on fractured surfaces matching[J]. Journal of Image and Graphics, 2012, 17(10):1298-1304(in Chinese)(李群辉,周明全,耿国华.断裂面匹配的破碎刚体复原[J].中国图象图形学报, 2012, 17(10):1298-1304)
[9]LiuJun,ZhouMingquan,GengGuohua,etal.Fragments splicing method for terra-cotta figures of Qin Dynasty based on contoursandfracturesurfacesmatching[J].ComputerEngineering, 2014, 40(1):181-185+190(in Chinese)(刘军,周明全,耿国华,等.基于轮廓与断面匹配的秦俑碎片拼接方法[J].计算机工程, 2014, 40(1):181-185+190)
[10]Li Shanshan, Geng Guohua, Zhou Mingquan, et al. Interactive reassembly of fractured fragments based on surface adjacency constraint[J]. Journal of Computer-Aided Design&Computer Graphics, 2016, 28(6):924-931(in Chinese)(李姗姗,耿国华,周明全,等.基于表面邻接约束的交互式文物碎片重组[J].计算机辅助设计与图形学学报,2016,28(6):924-931)
[11]Zhou Mingquan, Yuan Jie, Geng Guohua, et al. Interactive reassemblyoffracturedfragmentsbasedonfeaturepointsof contour line[J]. Optics and Precision Engineering, 2017, 25(6):1597-1606(in Chinese)(周明全,袁洁,耿国华,等.基于轮廓线特征点的交互式文物拼接[J].光学精密工程, 2017, 25(6):1597-1606)
[12]YuanJie,ZhouMingquan,GengGuohua,etal.Heritagedebris splicing algorithm based on contour line two-way distance field[J].Computer Engineering, 2018, 44(6):207-212+218(in Chinese)(袁洁,周明全,耿国华,等.基于轮廓线双向距离场的文物碎片拼接算法[J].计算机工程, 2018, 44(6):207-212+218)
[13]Zhong Y. Intrinsic shape signatures:a shape descriptor for 3D objectrecognition[C]//Proceedingsofthe12thInternational ConferenceonComputerVisionWorkshops.LosAlamitos:IEEE Computer Society Press, 2009:689-696
[14]Rusu R B, Blodow N, Beetz M. Fast point feature histograms(FPFH)for 3D registration[C]//Proceedings of the IEEE International Conference on Robotics and Automation. Los Alamitos:IEEE Computer Society Press, 2009:1848-1853
[15]Fischler M A, Bolles R C. Random sample consensus:a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM, 1981,24(6):381-395
[16]BeslPJ,MckayHD.Amethodforregistrationof3-D shapes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992, 14(2):239-256
[17]Johnson A E, Hebert M. Using spin images for efficient object recognition in cluttered 3D scenes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1999, 21(5):433-449
[18]BelongieS,MalikJ.Matchingwithshapecontexts[C]//Proceedings of the IEEE Workshop on Content-Based Access of Image and Video Libraries. Los Alamitos:IEEE Computer Society Press, 2000:20
[19]Sipiran I, Bustos B. Harris 3D:a robust extension of the Harris operator for interest point detection on 3D meshes[J]. The Visual Computer, 2011, 27(11):963-976
[20]Rusu R B, Marton Z C, Blodow N, et al. Learning informative pointclassesfortheacquisitionofobjectmodelmaps[C]//Proceedings of the 10th International Conference on Control,Automation, Robotics andVision. LosAlamitos:IEEE Computer Society Press, 2008:643-650
[21]TombariF,SaltiS,StefanoLD.Uniquesignaturesofhistograms for local surface description[C]//Proceedings of the 11th EuropeanConferenceonComputerVision.Heidelberg:Springer, 2010, Part III:356-369
[22]Chen Z, Czarnuch S, Smith A, et al. Performance evaluation of 3D keypointsand descriptors[C]//Proceedings ofInternationalSymposium on Visual Computing. Heidelberg:Springer, 2016:410-420
[23]Yuan Jie, Zhou Mingquan, Geng Guohua, et al. Automatic reassembly of fractured fragments using morse topological features[J].Acta Automatica Sinica, 2018, 44(8):1486-1495(in Chinese)(袁洁,周明全,耿国华,等.基于Morse-Smale拓扑特征的文物碎片拼接算法[J].自动化学报, 2018, 44(8):1486-1495)