The Experts below are selected from a list of 33705 Experts worldwide ranked by ideXlab platform
Kenji Shimada - One of the best experts on this subject based on the ideXlab platform.
-
fully automated hex dominant mesh generation with directionality control via packing rectangular solid cells
International Journal for Numerical Methods in Engineering, 2003Co-Authors: Soji Yamakawa, Kenji ShimadaAbstract:A new fully automatic hex-dominant mesh generation technique of an arbitrary 3D Geometric Domain is presented herein. The proposed method generates a high-quality hex-dominant mesh by: (1) controlling the directionality of the output hex-dominant mesh; and (2) avoiding ill-shaped elements induced by nodes located too closely to each other. The proposed method takes a 3D Geometric Domain as input and creates a hex-dominant mesh consisting mostly of hexahedral elements, with additional prism and tetrahedral elements. Rectangular solid cells are packed on the boundary of and inside the input Domain to obtain ideal node locations for a hex-dominant mesh. Each cell has a potential energy field that mimics a body-centred cubic (BCC) structure (seen in natural substances such as NaCl) and the cells are moved to stable positions by a physically based simulation. The simulation mimics the formation of a crystal pattern so that the centres of the cells provide ideal node locations for a hex-dominant mesh. Via the advancing front method, the centres of the packed cells are then connected to form a tetrahedral mesh, and this is converted to a hex-dominant mesh by merging some of the tetrahedrons. Copyright © 2003 John Wiley & Sons, Ltd.
-
hex dominant mesh generation with directionality control via packing rectangular solid cells
Geometric Modeling and Processing, 2002Co-Authors: Soji Yamakawa, Kenji ShimadaAbstract:A new computational method that creates a hex-dominant mesh of an arbitrary 3D Geometric Domain is presented. The proposed method generates a high-quality hex-dominant mesh by: (1) controlling the directionality of the output hex-dominant mesh; and (2) avoiding ill-shaped elements induced by nodes located too closely to each other. The proposed method takes a 3D Geometric Domain as input and creates a hex-dominant mesh that consists of mostly hexahedral elements with additional prism elements and tetrahedral elements. The proposed method packs rectangular solid cells on the boundary of and inside the input Domain to obtain ideal node locations for a hex-dominant mesh. Each cell has a potential energy field that mimics a body centered cubic (BCC) structure, and the cells are moved to stable positions by a physically-based simulation. The simulation mimics the formation of a crystal pattern so that the centers of the cells give ideal node locations for a hex-dominant mesh. The Domain is then meshed into a tetrahedral mesh by the advancing front method, and finally the tetrahedral mesh is converted to a hex-dominant mesh by merging some tetrahedrons.
-
high quality anisotropic tetrahedral mesh generation via ellipsoidal bubble packing
IMR, 2000Co-Authors: Soji Yamakawa, Kenji ShimadaAbstract:This paper presents a new computational method for anisotropic tetrahedral meshing that (1) can control shapes of the elements by an arbitrary anisotropy function, and (2) can avoid ill-shaped elements induced from poorly distributed node locations. Our method creates a tetrahedral mesh in two steps. First our method obtains node locations through a physically based particle simulation, which we call 'bubble packing.' Ellipsoidal bubbles are closely packed on the boundary and inside a Geometric Domain, and nodes are placed at the centers of the bubbles. Our method then connects the nodes to create a tet mesh by the advancing front method. Experimental results show that our method can create a high quality anisotropic tetrahedral mesh that conforms well to the input anisotropy.
Soji Yamakawa - One of the best experts on this subject based on the ideXlab platform.
-
fully automated hex dominant mesh generation with directionality control via packing rectangular solid cells
International Journal for Numerical Methods in Engineering, 2003Co-Authors: Soji Yamakawa, Kenji ShimadaAbstract:A new fully automatic hex-dominant mesh generation technique of an arbitrary 3D Geometric Domain is presented herein. The proposed method generates a high-quality hex-dominant mesh by: (1) controlling the directionality of the output hex-dominant mesh; and (2) avoiding ill-shaped elements induced by nodes located too closely to each other. The proposed method takes a 3D Geometric Domain as input and creates a hex-dominant mesh consisting mostly of hexahedral elements, with additional prism and tetrahedral elements. Rectangular solid cells are packed on the boundary of and inside the input Domain to obtain ideal node locations for a hex-dominant mesh. Each cell has a potential energy field that mimics a body-centred cubic (BCC) structure (seen in natural substances such as NaCl) and the cells are moved to stable positions by a physically based simulation. The simulation mimics the formation of a crystal pattern so that the centres of the cells provide ideal node locations for a hex-dominant mesh. Via the advancing front method, the centres of the packed cells are then connected to form a tetrahedral mesh, and this is converted to a hex-dominant mesh by merging some of the tetrahedrons. Copyright © 2003 John Wiley & Sons, Ltd.
-
hex dominant mesh generation with directionality control via packing rectangular solid cells
Geometric Modeling and Processing, 2002Co-Authors: Soji Yamakawa, Kenji ShimadaAbstract:A new computational method that creates a hex-dominant mesh of an arbitrary 3D Geometric Domain is presented. The proposed method generates a high-quality hex-dominant mesh by: (1) controlling the directionality of the output hex-dominant mesh; and (2) avoiding ill-shaped elements induced by nodes located too closely to each other. The proposed method takes a 3D Geometric Domain as input and creates a hex-dominant mesh that consists of mostly hexahedral elements with additional prism elements and tetrahedral elements. The proposed method packs rectangular solid cells on the boundary of and inside the input Domain to obtain ideal node locations for a hex-dominant mesh. Each cell has a potential energy field that mimics a body centered cubic (BCC) structure, and the cells are moved to stable positions by a physically-based simulation. The simulation mimics the formation of a crystal pattern so that the centers of the cells give ideal node locations for a hex-dominant mesh. The Domain is then meshed into a tetrahedral mesh by the advancing front method, and finally the tetrahedral mesh is converted to a hex-dominant mesh by merging some tetrahedrons.
-
high quality anisotropic tetrahedral mesh generation via ellipsoidal bubble packing
IMR, 2000Co-Authors: Soji Yamakawa, Kenji ShimadaAbstract:This paper presents a new computational method for anisotropic tetrahedral meshing that (1) can control shapes of the elements by an arbitrary anisotropy function, and (2) can avoid ill-shaped elements induced from poorly distributed node locations. Our method creates a tetrahedral mesh in two steps. First our method obtains node locations through a physically based particle simulation, which we call 'bubble packing.' Ellipsoidal bubbles are closely packed on the boundary and inside a Geometric Domain, and nodes are placed at the centers of the bubbles. Our method then connects the nodes to create a tet mesh by the advancing front method. Experimental results show that our method can create a high quality anisotropic tetrahedral mesh that conforms well to the input anisotropy.
Kufoniyi O. - One of the best experts on this subject based on the ideXlab platform.
-
Spatial coincidence modelling, automated database updating and data consistency in vector GIS.
Kufoniyi, 1995Co-Authors: Kufoniyi O.Abstract:This thesis presents formal approaches for automated database updating and consistency control in vector- structured spatial databases. To serve as a framework, a conceptual data model is formalized for the representation of geo-data from multiple map layers in which a map layer denotes a set of terrain objects of the same mapping context, e.g., cadastral, soil mapping, etc. The necessity for a generalised model arises from the frequent requirement in spatial analysis and planning for a Geometric integration of several different views of the world, whereas most existing data models were designed from the perspective of a "single application" leading to ad hoc and repeated overlay computations (during query processing) when dealing with an integrated analysis. An alternative model is therefore proposed in this thesis for the Geometric integration of geo-data from multiple map layers. The proposed model is an object-based, query-oriented 2.5D data model for multi-valued vector maps (DMMVM).A multi-valued vector map refers to the vector-based representation of terrain objects from multiple map layers whereby two objects of the same Geometric type may be spatially coincident. Two objects of the same type are spatially coincident if they (partially) overlap in space. In this model, positions of objects are defined in a 3D metric space but embedded in 2D topologic space. The model is based on the 2D formal data structure (FDS) for singlevalued vector maps.Terrain objects play a central role in the terrain description; each object has a thematic component and a Geometric component. In the thematic Domain, the objects can be grouped into thematic classes in which each class has a specific attribute structure, and in the Geometric Domain the object types (points, lines and areas) are distinguished for a 2D or 2.5D terrain description.A Geometric data type -- the m-dimensional container, or simply m-container, where m ε{0,1,2} -- is then introduced to model spatial coincidence among objects of the same Geometric type. By introducing the container data type, overlapping sections across the layers are uniquely identified such that they have their own individual Geometric data and nonspatial data, apart from those inherited from the overlapping objects; they can then be maintained and manipulated by the DBMS just like single objects. Using graph theory as a mathematical tool, the three container types are then represented by the topologic primitives arc and node. A node defines one 0-container andlor the beginning or end of an arc, while an arc defines (part of) one 1-container andlor (part of) the boundary of a 2-container. The arc is defined by one start node and one end node, and a node is defined by a coordinate triplet X,Y,Z. A flexible integration of the model with a DTM is also presented in the thesis, using an edge-based TIN. Two primitives of the edge-based TIN (edge and vertex) are added to the data types of the DMMVM to define the integrated model.Research and development on the updating of geo-information have been confined mainly to the aspects of data collection and change detection, with little emphasis on the corresponding automated propagation of the updating in the database in a consistent manner. To address the latter aspect, procedures are formulated in the thesis for a consistent automated updating of a vector-structured database, using the DMMVM as a framework. Algorithms are provided for the automated update propagation such that topology is automatically updated by the system, while maintaining structural and semantic consistency. This will improve on the current practice in operational systems, which usually requires a delayed reconstruction of topology whenever there is a Geometric change in the database. Algorithms are developed for the insertion, deletion or modification of each of the eight data types (area, line, point, 2- container, 1-container, 0-container, arc and node) in the DMMVM. The human operator interacts with the system at the object-level, while the system propagates the update. The topology of the database is updated dynamically by the system by evaluating, using computational geometry, the topologic relationship between the new primitive (are or node) of an object and the existing primitives in the database. The type of relationship detected will then activate the relevant consistency rule (including update propagation) to validate the topology and consistency of the database. The system alerts the human operator if it is not possible for it to resolve the inconsistency.To enforce data consistency during Geometric updating of the database, consistency rules are defined to ensure structural consistency, while a monitoring strategy is formulated for semantic (application-dependent, topologic) constraints. In both cases, topology plays the central role as an "alerter" of constraint violations. Thus the possible topologic relationships among the three elementary object types (area, line and point), and among the Geometric primitives (arc and node) in the DMMVM are formalised and algorithms are defined for detecting the occurrence of any of the elementary relationships for any object combination. Then the consistency constraints can be translated to topologic relationships and stored in the database as events, and the corresponding responses of the system to enforce consistency can be defined as actions, thus giving a rule-based procedure (using the if event then action convention) for the management of data consistency in spatial databases.The DMMVM was translated into a relational database structure and an object-oriented database structure to facilitate implementation in a variety of systems. The object-oriented data structure and the consistency rules and algorithms were tested experimentally using Postgres, an extended relational database management system. Data were acquired using the Kork digital mapping system, on a Planicomp C120 photogrammetric stereoplotter equipped with a Zeiss Videomap and a Calcomp drawing board digitizer.The thesis concludes with an evaluation of the proposed model and an outline of areas requiring further investigations
-
Spatial coincidence modelling, automated database updating and data consistency in vector GIS
Kufoniyi, 1995Co-Authors: Kufoniyi O.Abstract:<p>This thesis presents formal approaches for automated database updating and consistency control in vector- structured spatial databases. To serve as a framework, a conceptual data model is formalized for the representation of geo-data from multiple map layers in which a map layer denotes a set of terrain objects of the same mapping context, e.g., cadastral, soil mapping, etc. The necessity for a generalised model arises from the frequent requirement in spatial analysis and planning for a Geometric integration of several different views of the world, whereas most existing data models were designed from the perspective of a "single application" leading to ad hoc and repeated overlay computations (during query processing) when dealing with an integrated analysis. An alternative model is therefore proposed in this thesis for the Geometric integration of geo-data from multiple map layers. The proposed model is an object-based, query-oriented 2.5D data model for multi-valued vector maps (DMMVM).<p>A multi-valued vector map refers to the vector-based representation of terrain objects from multiple map layers whereby two objects of the same Geometric type may be spatially coincident. Two objects of the same type are spatially coincident if they (partially) overlap in space. In this model, positions of objects are defined in a 3D metric space but embedded in 2D topologic space. The model is based on the 2D formal data structure (FDS) for singlevalued vector maps.<p>Terrain objects play a central role in the terrain description; each object has a thematic component and a Geometric component. In the thematic Domain, the objects can be grouped into thematic classes in which each class has a specific attribute structure, and in the Geometric Domain the object types (points, lines and areas) are distinguished for a 2D or 2.5D terrain description.<p>A Geometric data type -- the m-dimensional container, or simply m-container, where m ε{0,1,2} -- is then introduced to model spatial coincidence among objects of the same Geometric type. By introducing the container data type, overlapping sections across the layers are uniquely identified such that they have their own individual Geometric data and nonspatial data, apart from those inherited from the overlapping objects; they can then be maintained and manipulated by the DBMS just like single objects. Using graph theory as a mathematical tool, the three container types are then represented by the topologic primitives arc and node. A node defines one 0-container andlor the beginning or end of an arc, while an arc defines (part of) one 1-container andlor (part of) the boundary of a 2-container. The arc is defined by one start node and one end node, and a node is defined by a coordinate triplet X,Y,Z. A flexible integration of the model with a DTM is also presented in the thesis, using an edge-based TIN. Two primitives of the edge-based TIN (edge and vertex) are added to the data types of the DMMVM to define the integrated model.<p>Research and development on the updating of geo-information have been confined mainly to the aspects of data collection and change detection, with little emphasis on the corresponding automated propagation of the updating in the database in a consistent manner. To address the latter aspect, procedures are formulated in the thesis for a consistent automated updating of a vector-structured database, using the DMMVM as a framework. Algorithms are provided for the automated update propagation such that topology is automatically updated by the system, while maintaining structural and semantic consistency. This will improve on the current practice in operational systems, which usually requires a delayed reconstruction of topology whenever there is a Geometric change in the database. Algorithms are developed for the insertion, deletion or modification of each of the eight data types (area, line, point, 2- container, 1-container, 0-container, arc and node) in the DMMVM. The human operator interacts with the system at the object-level, while the system propagates the update. The topology of the database is updated dynamically by the system by evaluating, using computational geometry, the topologic relationship between the new primitive (are or node) of an object and the existing primitives in the database. The type of relationship detected will then activate the relevant consistency rule (including update propagation) to validate the topology and consistency of the database. The system alerts the human operator if it is not possible for it to resolve the inconsistency.<p>To enforce data consistency during Geometric updating of the database, consistency rules are defined to ensure structural consistency, while a monitoring strategy is formulated for semantic (application-dependent, topologic) constraints. In both cases, topology plays the central role as an "alerter" of constraint violations. Thus the possible topologic relationships among the three elementary object types (area, line and point), and among the Geometric primitives (arc and node) in the DMMVM are formalised and algorithms are defined for detecting the occurrence of any of the elementary relationships for any object combination. Then the consistency constraints can be translated to topologic relationships and stored in the database as <em>events,</em> and the corresponding responses of the system to enforce consistency can be defined as actions, thus giving a rule-based procedure (using the if <em>event</em> then <em>action</em> convention) for the management of data consistency in spatial databases.<p>The DMMVM was translated into a relational database structure and an object-oriented database structure to facilitate implementation in a variety of systems. The object-oriented data structure and the consistency rules and algorithms were tested experimentally using Postgres, an extended relational database management system. Data were acquired using the Kork digital mapping system, on a Planicomp C120 photogrammetric stereoplotter equipped with a Zeiss Videomap and a Calcomp drawing board digitizer.<p>The thesis concludes with an evaluation of the proposed model and an outline of areas requiring further investigations
Zohaib Atiq Khan - One of the best experts on this subject based on the ideXlab platform.
-
efficient extraction of pore networks from massive tomograms via Geometric Domain decomposition
Advances in Water Resources, 2020Co-Authors: Zohaib Atiq Khan, Ali Elkamel, Jeff T GostickAbstract:Abstract Image-based modelling of porous media to study the transport and reaction processes has become an essential tool. The availability of increasingly large image datasets at high resolution creates a need to develop algorithms that can process massive size images at a low computational cost. This study presents an efficient workflow to extract pore networks form large size porous Domains using a watershed segmentation with Geometrical Domain decomposition. The method subdivides a porous image into smaller overlapping subDomains and performs a watershed segmentation on each subDomains in parallel or serial modes of operation to save CPU time or memory RAM, respectively. The computational performance of the algorithm was analyzed on a large size image and found to consume 50 percent less memory and upto 7 times less CPU time than the standard watershed implementation. Pore networks of four massive digital rock images were extracted and the the effective permeability predicted by the networks agreed well with previously investigated values illustrating the accuracy of the method. An additional application of this methods, taking advantage of the reduced computational cost, is the upgrading of low-resolution image. It was found that that increasing the resolution of a coarse image leads to more accurate predictions by helping the watershed segmenation prouduce a more faithful pore network model. The developed algorithm is implemented in Python, and included in the open source project PoreSpy. It uses highly optimized and efficient modules such as Dask and Numba to obtain the maximum performance. The Domain decomposition approach used here will also lend itself well to processing on distributed memory clusters, enabling the processing of even larger porous Domains.
Jeff T Gostick - One of the best experts on this subject based on the ideXlab platform.
-
efficient extraction of pore networks from massive tomograms via Geometric Domain decomposition
Advances in Water Resources, 2020Co-Authors: Zohaib Atiq Khan, Ali Elkamel, Jeff T GostickAbstract:Abstract Image-based modelling of porous media to study the transport and reaction processes has become an essential tool. The availability of increasingly large image datasets at high resolution creates a need to develop algorithms that can process massive size images at a low computational cost. This study presents an efficient workflow to extract pore networks form large size porous Domains using a watershed segmentation with Geometrical Domain decomposition. The method subdivides a porous image into smaller overlapping subDomains and performs a watershed segmentation on each subDomains in parallel or serial modes of operation to save CPU time or memory RAM, respectively. The computational performance of the algorithm was analyzed on a large size image and found to consume 50 percent less memory and upto 7 times less CPU time than the standard watershed implementation. Pore networks of four massive digital rock images were extracted and the the effective permeability predicted by the networks agreed well with previously investigated values illustrating the accuracy of the method. An additional application of this methods, taking advantage of the reduced computational cost, is the upgrading of low-resolution image. It was found that that increasing the resolution of a coarse image leads to more accurate predictions by helping the watershed segmenation prouduce a more faithful pore network model. The developed algorithm is implemented in Python, and included in the open source project PoreSpy. It uses highly optimized and efficient modules such as Dask and Numba to obtain the maximum performance. The Domain decomposition approach used here will also lend itself well to processing on distributed memory clusters, enabling the processing of even larger porous Domains.