Bjorn Van Genechten | KU Leuven (original) (raw)
Uploads
Papers by Bjorn Van Genechten
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Nov 13, 2017
Point cloud segmentation is a crucial step in scene understanding and interpretation. The goal is... more Point cloud segmentation is a crucial step in scene understanding and interpretation. The goal is to decompose the initial data into sets of workable clusters with similar properties. Additionally, it is a key aspect in the automated procedure from point cloud data to BIM. Current approaches typically only segment a single type of primitive such as planes or cylinders. Also, current algorithms suffer from oversegmenting the data and are often sensor or scene dependent. In this work, a method is presented to automatically segment large unstructured point clouds of buildings. More specifically, the segmentation is formulated as a graph optimisation problem. First, the data is oversegmented with a greedy octree-based region growing method. The growing is conditioned on the segmentation of planes as well as smooth surfaces. Next, the candidate clusters are represented by a Conditional Random Field after which the most likely configuration of candidate clusters is computed given a set of local and contextual features. The experiments prove that the used method is a fast and reliable framework for unstructured point cloud segmentation. Processing speeds up to 40,000 points per second are recorded for the region growing. Additionally, the recall and precision of the graph clustering is approximately 80%. Overall, nearly 22% of oversegmentation is reduced by clustering the data. These clusters will be classified and used as a basis for the reconstruction of BIM models.
Wiadomości Konserwatorskie, 2009
Virtual Systems and MultiMedia, 2005
Documenting cultural heritage is an important step towards the conservation and restoration of mo... more Documenting cultural heritage is an important step towards the conservation and restoration of monuments. 'Precise documentation' in conservation works, deals with the use of techniques capable of acquiring geometry as well as texture information with high precision and integrity. Only recently orthophoto's have been accepted and are used as a standard tool in archeological and cultural heritage documentation. By analyzing two existing software packages for orthophotograph production, we conclude that standard automatic surface modeling techniques as well as laser scanning do not fulfill the high needs of close range architectural orthophotography because of the large amount of discontinuities in the surface. This paper presents ongoing research that combines the use of laser scanning, photogrammetry and image processing to overcome these problems. The major problem lies in achieving automation and efficiency together with robustness and reliability so that the algorithms and methods can be used for a successful and economically feasible implementation in practice.
Optics and Lasers in Engineering, Mar 1, 2009
Safety and stability of historical structures are of key importance when dealing with a restorati... more Safety and stability of historical structures are of key importance when dealing with a restoration project. The stability of masonry vaults very much depends on its overall geometry and the analysis results strongly depend on the accuracy in which this geometry can be measured in practice. To obtain an accurate estimate of the geometry of the vaults, 3D-laser scanning was performed. Based on the 3D-point cloud, a 3D-model of the vaults allows determining the lines of thrust in the structure as well as the reaction forces (application point and magnitude) at the abutments. This contribution gives a critical appraisal of the process used in the preliminary research phase: gathering the point clouds, 3D-modelling, structural analysis calculating thrust lines and the consolidation requirements that are obtained. The focus is on the added value of the methodology, its applicability, cost-effectiveness, overall advantages and drawbacks.
This paper describes the progress of an ongoing project called '3DRISKMAPPING&am... more This paper describes the progress of an ongoing project called '3DRISKMAPPING'. The purpose of this project is to create an e-learning platform for the use of 3D terrestrial laser scanning techniques for risk characterization of our built environment. We describe the general structure of the tutorial and its didactic approach. Points of focus are described in more detail as to give an idea of the final outcome of the project
When monuments are seriously damaged, or completely destroyed, the amount and quality of document... more When monuments are seriously damaged, or completely destroyed, the amount and quality of documentation becomes highly important. Nowadays orthophoto's have become a standard tool in archeological and in cultural heritage documentation. By analyzing two existing software packages that are used to create orthophotographs today, we find that standard automatic surface modeling techniques as well as laser scanning do not fulfill the high needs of close range architectural orthophotography because of the large amount of discontinuities in the surface. In this paper we plan to combine the use of laser scanning, photogrammetry and image processing to overcome these problems. The major problem lies in achieving automation and efficiency together with robustness and reliability so that the algorithms and methods can be used for a successful and economically feasible implementation in practice. I
ABSTRACT Thanks to its speed and accuracy, terrestrial laser scanning is gaining importance in th... more ABSTRACT Thanks to its speed and accuracy, terrestrial laser scanning is gaining importance in the field of architectural and cultural heritage. Laser scanners are field-of-view devices that offer a dense point-wise sampling of an object"s geometry. The first part of this article addresses the question whether laser scanning produces sufficiently accurate results to be used for deformation monitoring of historical structures. A test setup using a masonry arch was prepared in which different scenarios were simulated. The structure was scanned at different times while moving one of its legs and the influence of different scan resolutions on the accuracy and the detectability of deformations were tested. Using a different setup, the influence on accuracy of different targets configurations, in particular orientations and distances to the scanner were tested. In a second part, two prototype algorithms are presented for the automatic generation of cross sections based on point clouds. The first method is based on point splats. Point splats are an alternative for meshes for the surface-like reconstruction of point clouds. Their main advantages are the low computational cost and the ability to give an accurate representation of sharp edges and details. The cross sections obtained from our algorithm are compared to similar cross sections obtained from a mesh based surface representation. The second proposed method segments a sliced point cloud into outlier-free clusters based on techniques from robust statistics. The cluster is grown by fitting a parametric model and computing the residuals using a moving least squares approach. This procedure generates a piecewise smooth cross-section with sharp discontinuity representations. In order to prove the usability of the algorithm, the cross sections obtained from this algorithm are compared to similar cross sections obtained from a mesh based surface representation.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2018
The automated reconstruction of Building Information Modeling (BIM) objects from point cloud data... more The automated reconstruction of Building Information Modeling (BIM) objects from point cloud data is still ongoing research. A key aspect is the creation of accurate wall geometry as it forms the basis for further reconstruction of objects in a BIM. After segmenting and classifying the initial point cloud, the labelled segments are processed and the wall topology is reconstructed. However, the preocedure is challenging due to noise, occlusions and the complexity of the input data. In this work, a method is presented to automatically reconstruct consistent wall geometry from point clouds. More specifically, the use of room information is proposed to aid the wall topology creation. First, a set of partial walls is constructed based on classified planar primitives. Next, the rooms are identified using the retrieved wall information along with the floors and ceilings. The wall topology is computed by the intersection of the partial walls conditioned on the room information. The final wall geometry is defined by creating IfcWallStandardCase objects conform the IFC4 standard. The result is a set of walls according to the as-built conditions of a building. The experiments prove that the used method is a reliable framework for wall reconstruction from unstructured point cloud data. Also, the implementation of room information reduces the rate of false positives for the wall topology. Given the walls, ceilings and floors, 94% of the rooms is correctly identified. A key advantage of the proposed method is that it deals with complex rooms and is not bound to single storeys.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017
Semantically rich three dimensional models such as Building Information Models (BIMs) are increas... more Semantically rich three dimensional models such as Building Information Models (BIMs) are increasingly used in digital heritage. They provide the required information to varying stakeholders during the different stages of the historic buildings life cyle which is crucial in the conservation process. The creation of as-built BIM models is based on point cloud data. However, manually interpreting this data is labour intensive and often leads to misinterpretations. By automatically classifying the point cloud, the information can be proccesed more effeciently. A key aspect in this automated scan-to-BIM process is the classification of building objects. In this research we look to automatically recognise elements in existing buildings to create compact semantic information models. Our algorithm efficiently extracts the main structural components such as floors, ceilings, roofs, walls and beams despite the presence of significant clutter and occlusions. More specifically, Support Vector Machines (SVM) are proposed for the classification. The algorithm is evaluated using real data of a variety of existing buildings. The results prove that the used classifier recognizes the objects with both high precision and recall. As a result, entire data sets are reliably labelled at once. The approach enables experts to better document and process heritage assets.
ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2016
With the increasing popularity of as-built building models for the architectural, engineering and... more With the increasing popularity of as-built building models for the architectural, engineering and construction (AEC) industry, the demand for highly accurate and dense point cloud data is rising. The current data acquisition methods are labour intensive and time consuming. In order to compete with indoor mobile mapping systems (IMMS), surveyors are now opting to use terrestrial laser scanning as a standalone solution. However, there is uncertainty about the accuracy of this approach. The emphasis of this paper is to determine the scope for which terrestrial laser scanners can be used without additional control. Multiple real life test cases are evaluated in order to identify the boundaries of this technique. Furthermore, this research presents a mathematical prediction model that provides an indication of the data accuracy given the project dimensions. This will enable surveyors to make informed discussions about the employability of terrestrial laser scanning without additional con...
In recent years, 3D laser scanners have gained interest, especially in 3D reconstruction and repr... more In recent years, 3D laser scanners have gained interest, especially in 3D reconstruction and reproduction, mainly because of their capability to record huge numbers of points with high accuracy in a relatively short period of time. Especially in the field of deformation monitoring for analyzing the structural stability of built heritage, laser scanners can provide a much denser and thus better representation of the deformations taking place over time. The single-point precision of modern mid-range laser scanners varies from 5 to 25mm depending on the laser and the measuring principle used, but also depending on the measurement conditions and the object surface to be scanned. Comparing this precision to that of traditional surveying techniques used for deformation monitoring such as total stations or contact sensors, the precision of a laser scanner is approximately 1 order lower in magnitude. The main disadvantage of these traditional techniques is that they only offer single-point ...
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Nov 13, 2017
Point cloud segmentation is a crucial step in scene understanding and interpretation. The goal is... more Point cloud segmentation is a crucial step in scene understanding and interpretation. The goal is to decompose the initial data into sets of workable clusters with similar properties. Additionally, it is a key aspect in the automated procedure from point cloud data to BIM. Current approaches typically only segment a single type of primitive such as planes or cylinders. Also, current algorithms suffer from oversegmenting the data and are often sensor or scene dependent. In this work, a method is presented to automatically segment large unstructured point clouds of buildings. More specifically, the segmentation is formulated as a graph optimisation problem. First, the data is oversegmented with a greedy octree-based region growing method. The growing is conditioned on the segmentation of planes as well as smooth surfaces. Next, the candidate clusters are represented by a Conditional Random Field after which the most likely configuration of candidate clusters is computed given a set of local and contextual features. The experiments prove that the used method is a fast and reliable framework for unstructured point cloud segmentation. Processing speeds up to 40,000 points per second are recorded for the region growing. Additionally, the recall and precision of the graph clustering is approximately 80%. Overall, nearly 22% of oversegmentation is reduced by clustering the data. These clusters will be classified and used as a basis for the reconstruction of BIM models.
Wiadomości Konserwatorskie, 2009
Virtual Systems and MultiMedia, 2005
Documenting cultural heritage is an important step towards the conservation and restoration of mo... more Documenting cultural heritage is an important step towards the conservation and restoration of monuments. 'Precise documentation' in conservation works, deals with the use of techniques capable of acquiring geometry as well as texture information with high precision and integrity. Only recently orthophoto's have been accepted and are used as a standard tool in archeological and cultural heritage documentation. By analyzing two existing software packages for orthophotograph production, we conclude that standard automatic surface modeling techniques as well as laser scanning do not fulfill the high needs of close range architectural orthophotography because of the large amount of discontinuities in the surface. This paper presents ongoing research that combines the use of laser scanning, photogrammetry and image processing to overcome these problems. The major problem lies in achieving automation and efficiency together with robustness and reliability so that the algorithms and methods can be used for a successful and economically feasible implementation in practice.
Optics and Lasers in Engineering, Mar 1, 2009
Safety and stability of historical structures are of key importance when dealing with a restorati... more Safety and stability of historical structures are of key importance when dealing with a restoration project. The stability of masonry vaults very much depends on its overall geometry and the analysis results strongly depend on the accuracy in which this geometry can be measured in practice. To obtain an accurate estimate of the geometry of the vaults, 3D-laser scanning was performed. Based on the 3D-point cloud, a 3D-model of the vaults allows determining the lines of thrust in the structure as well as the reaction forces (application point and magnitude) at the abutments. This contribution gives a critical appraisal of the process used in the preliminary research phase: gathering the point clouds, 3D-modelling, structural analysis calculating thrust lines and the consolidation requirements that are obtained. The focus is on the added value of the methodology, its applicability, cost-effectiveness, overall advantages and drawbacks.
This paper describes the progress of an ongoing project called '3DRISKMAPPING&am... more This paper describes the progress of an ongoing project called '3DRISKMAPPING'. The purpose of this project is to create an e-learning platform for the use of 3D terrestrial laser scanning techniques for risk characterization of our built environment. We describe the general structure of the tutorial and its didactic approach. Points of focus are described in more detail as to give an idea of the final outcome of the project
When monuments are seriously damaged, or completely destroyed, the amount and quality of document... more When monuments are seriously damaged, or completely destroyed, the amount and quality of documentation becomes highly important. Nowadays orthophoto's have become a standard tool in archeological and in cultural heritage documentation. By analyzing two existing software packages that are used to create orthophotographs today, we find that standard automatic surface modeling techniques as well as laser scanning do not fulfill the high needs of close range architectural orthophotography because of the large amount of discontinuities in the surface. In this paper we plan to combine the use of laser scanning, photogrammetry and image processing to overcome these problems. The major problem lies in achieving automation and efficiency together with robustness and reliability so that the algorithms and methods can be used for a successful and economically feasible implementation in practice. I
ABSTRACT Thanks to its speed and accuracy, terrestrial laser scanning is gaining importance in th... more ABSTRACT Thanks to its speed and accuracy, terrestrial laser scanning is gaining importance in the field of architectural and cultural heritage. Laser scanners are field-of-view devices that offer a dense point-wise sampling of an object"s geometry. The first part of this article addresses the question whether laser scanning produces sufficiently accurate results to be used for deformation monitoring of historical structures. A test setup using a masonry arch was prepared in which different scenarios were simulated. The structure was scanned at different times while moving one of its legs and the influence of different scan resolutions on the accuracy and the detectability of deformations were tested. Using a different setup, the influence on accuracy of different targets configurations, in particular orientations and distances to the scanner were tested. In a second part, two prototype algorithms are presented for the automatic generation of cross sections based on point clouds. The first method is based on point splats. Point splats are an alternative for meshes for the surface-like reconstruction of point clouds. Their main advantages are the low computational cost and the ability to give an accurate representation of sharp edges and details. The cross sections obtained from our algorithm are compared to similar cross sections obtained from a mesh based surface representation. The second proposed method segments a sliced point cloud into outlier-free clusters based on techniques from robust statistics. The cluster is grown by fitting a parametric model and computing the residuals using a moving least squares approach. This procedure generates a piecewise smooth cross-section with sharp discontinuity representations. In order to prove the usability of the algorithm, the cross sections obtained from this algorithm are compared to similar cross sections obtained from a mesh based surface representation.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2018
The automated reconstruction of Building Information Modeling (BIM) objects from point cloud data... more The automated reconstruction of Building Information Modeling (BIM) objects from point cloud data is still ongoing research. A key aspect is the creation of accurate wall geometry as it forms the basis for further reconstruction of objects in a BIM. After segmenting and classifying the initial point cloud, the labelled segments are processed and the wall topology is reconstructed. However, the preocedure is challenging due to noise, occlusions and the complexity of the input data. In this work, a method is presented to automatically reconstruct consistent wall geometry from point clouds. More specifically, the use of room information is proposed to aid the wall topology creation. First, a set of partial walls is constructed based on classified planar primitives. Next, the rooms are identified using the retrieved wall information along with the floors and ceilings. The wall topology is computed by the intersection of the partial walls conditioned on the room information. The final wall geometry is defined by creating IfcWallStandardCase objects conform the IFC4 standard. The result is a set of walls according to the as-built conditions of a building. The experiments prove that the used method is a reliable framework for wall reconstruction from unstructured point cloud data. Also, the implementation of room information reduces the rate of false positives for the wall topology. Given the walls, ceilings and floors, 94% of the rooms is correctly identified. A key advantage of the proposed method is that it deals with complex rooms and is not bound to single storeys.
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017
Semantically rich three dimensional models such as Building Information Models (BIMs) are increas... more Semantically rich three dimensional models such as Building Information Models (BIMs) are increasingly used in digital heritage. They provide the required information to varying stakeholders during the different stages of the historic buildings life cyle which is crucial in the conservation process. The creation of as-built BIM models is based on point cloud data. However, manually interpreting this data is labour intensive and often leads to misinterpretations. By automatically classifying the point cloud, the information can be proccesed more effeciently. A key aspect in this automated scan-to-BIM process is the classification of building objects. In this research we look to automatically recognise elements in existing buildings to create compact semantic information models. Our algorithm efficiently extracts the main structural components such as floors, ceilings, roofs, walls and beams despite the presence of significant clutter and occlusions. More specifically, Support Vector Machines (SVM) are proposed for the classification. The algorithm is evaluated using real data of a variety of existing buildings. The results prove that the used classifier recognizes the objects with both high precision and recall. As a result, entire data sets are reliably labelled at once. The approach enables experts to better document and process heritage assets.
ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2016
With the increasing popularity of as-built building models for the architectural, engineering and... more With the increasing popularity of as-built building models for the architectural, engineering and construction (AEC) industry, the demand for highly accurate and dense point cloud data is rising. The current data acquisition methods are labour intensive and time consuming. In order to compete with indoor mobile mapping systems (IMMS), surveyors are now opting to use terrestrial laser scanning as a standalone solution. However, there is uncertainty about the accuracy of this approach. The emphasis of this paper is to determine the scope for which terrestrial laser scanners can be used without additional control. Multiple real life test cases are evaluated in order to identify the boundaries of this technique. Furthermore, this research presents a mathematical prediction model that provides an indication of the data accuracy given the project dimensions. This will enable surveyors to make informed discussions about the employability of terrestrial laser scanning without additional con...
In recent years, 3D laser scanners have gained interest, especially in 3D reconstruction and repr... more In recent years, 3D laser scanners have gained interest, especially in 3D reconstruction and reproduction, mainly because of their capability to record huge numbers of points with high accuracy in a relatively short period of time. Especially in the field of deformation monitoring for analyzing the structural stability of built heritage, laser scanners can provide a much denser and thus better representation of the deformations taking place over time. The single-point precision of modern mid-range laser scanners varies from 5 to 25mm depending on the laser and the measuring principle used, but also depending on the measurement conditions and the object surface to be scanned. Comparing this precision to that of traditional surveying techniques used for deformation monitoring such as total stations or contact sensors, the precision of a laser scanner is approximately 1 order lower in magnitude. The main disadvantage of these traditional techniques is that they only offer single-point ...