Roof Edges Research Articles

Non-Contact Diagnostics of the Geometry of a Historic Wooden Building as an Element of Periodic Safety Assessment.

The article presents a method of non-invasive diagnostics of a historic wooden church, built in the 18th century. Over the hundreds of years of its use, changes in the geometry of the structure have been observed. This article presents the requirements of so-called architectural and geodetic survey and the method of using terrestrial laser scanning to create a three-dimensional solid model of an object. The diagnostic tests performed made it possible to perform analysis based on a so-called point cloud, which is a virtual representation of a real object. In order to determine the basic parameters of the building, the area and volume of all rooms were determined. It was found that the object exhibited deformations that cannot be explained solely as a result of imperfections during climb and normal wear and tear during operation. Therefore, the changes in shape were assessed in detail by means of an assessment of the verticality of the pillars supporting two levels of the church, the verticality of the walls, and the inconsistency of the floors, as well as the shape and horizontality of the roof edge. Additional InSAR and FEM tests of the object’s location on the ground allowed identification of the cause of the object deformation as the influence of inhomogeneous groundwater relations under the building. Without prophylactic measures, this deformation phenomenon can be expected to worsen. The tests described should therefore be considered as essential in subsequent diagnostic cycles and permit future extended numerical FEM analysis.

Roof Segmentation Towards Digital Twin Generation in LoD2+ Using Deep Learning

There is an increasing need for digital twins of cities and their base maps, 3D city models. Creating and updating these twins is not an easy task, so automating and streamlining the process is a field of active research. A significant part of the urban geometry is residential buildings and their roofs. Modeling of roofs for urban buildings can be divided into three main areas - building detection, roof recognition and building reconstruction. The building and roofs are segmented with the help of machine learning and image processing. Afterwards the extracted information is used to generate parametric models for the roofs using methods from computational geometry. The goal is to create correct virtual models of roofs belonging to many different types of buildings. In this study, a supervised deep learning approach is proposed for the segmentation of roof edges from a single orthophoto. The predicted features include the linear elements of roofs. The experiments show that, despite the small amount of training data, even in the presence of noise, the proposed method performs well on semantic segmentation of roofs with different shapes and complexities. The quality of the extracted roof elements for the test area is about 56% and 71% for mean intersection over union (IOU) and Dice metric scores, respectively.

In-situ measurement of structural performance of large-span air-supported dome under wind loads

To comprehensively understand the wind-induced response of air-supported membrane structure, a structural health monitoring system is designed and firstly installed on a pneumatic membrane sports hall with large span. Consisting of four sub-systems, this system was applied to perform a long-term in-situ monitoring. Structural vibration data of outer fabric membrane and steel cable tension strain were acquired during the passage of typhoon Lekima in 2019, as well as internal pressure and wind pressure posed on the structure.Natural frequency and damping ratio are identified based on acceleration data. First five modes are obtained with natural frequencies ranged from 0.279 Hz to 0.457 Hz. Damping ratios approximates 1.5%. Modal shapes are mainly along with lateral and vertical directions. Besides, monitoring results show that the internal pressure of the pneumatic sports hall slightly fluctuates with the structural geometry variation induced by external wind. Windward pressure and wind suction at roof top are primary factors affecting the deformation of outer fabric membrane and cable strain. Specifically, with wind intensity increasing the structure geometry changes from semispherical to conical shape in side view and is twisted by the moment induced via wind load. It is also found that cables at roof ridge area have larger strain than those close to roof edges.

Towards open-source LOD2 modelling using convolutional neural networks

The aim of this paper is to classify and segment roofs using vertical aerial imagery to generate three-dimensional (3D) models. Such models can be used, for example, to evaluate the rainfall runoff from properties for rainwater harvesting and in assessing solar energy and roof insulation options. Aerial orthophotos and building footprints are used to extract individual roofs and bounding boxes, which are then fed into one neural network for classification and then another for segmentation. The approach initially implements transfer learning on a pre-trained VGG16 model. The first step achieves an accuracy of 95.39% as well as a F1 score of 95%. The classified images are segmented using a fully convolutional network semantic segmentation model. The mask of the segmented roof planes is used to extract the coordinates of the roof edges and the nexus points using the Harris corner detector algorithm. The coordinates of the corners are then used to plot a 3D Level of Detail 2 (LOD2) representation of the building and the roof height is determined by calculating the maximum and minimum height of a Digital Surface Model LiDAR point cloud and known building height data. Subsequently the wireframe plot can be used to compute the roof area. This model achieved an accuracy of 80.2%, 96.1%, 96.0%, 85.1% and 91.1% for flat, hip, gable, cross-hip and mansard roofs, respectively.

Snake-Based Model for Automatic Roof Boundary Extraction in the Object Space Integrating a High-Resolution Aerial Images Stereo Pair and 3D Roof Models

The accelerated urban development over the last decades has made it necessary to update spatial information rapidly and constantly. Therefore, cities’ three-dimensional models have been widely used as a study base for various urban problems. However, although many efforts have been made to develop new building extraction methods, reliable and automatic extraction is still a major challenge for the remote sensing and computer vision communities, mainly due to the complexity and variability of urban scenes. This paper presents a method to extract building roof boundaries in the object space by integrating a high-resolution aerial images stereo pair, three-dimensional roof models reconstructed from light detection and ranging (LiDAR) data, and contextual information of the scenes involved. The proposed method focuses on overcoming three types of common problems that can disturb the automatic roof extraction in the urban environment: perspective occlusions caused by high buildings, occlusions caused by vegetation covering the roof, and shadows that are adjacent to the roofs, which can be misinterpreted as roof edges. For this, an improved Snake-based mathematical model is developed considering the radiometric and geometric properties of roofs to represent the roof boundary in the image space. A new approach for calculating the corner response and a shadow compensation factor was added to the model. The created model is then adapted to represent the boundaries in the object space considering a stereo pair of aerial images. Finally, the optimal polyline, representing a selected roof boundary, is obtained by optimizing the proposed Snake-based model using a dynamic programming (DP) approach considering the contextual information of the scene. The results showed that the proposed method works properly in boundary extraction of roofs with occlusion and shadows areas, presenting completeness and correctness average values above 90%, RMSE average values below 0.5 m for E and N components, and below 1 m for H component.

Popolni ortofoto v urbanih okoljih

True orthophoto is a better product than a classical orthophoto in urban areas, because buildings are depicted in the correct plane position, and the content around the building is also visible. The main goal of our research was to verify two approaches of a true orthophoto production: a) true orthophoto production based on digital terrain model and digital building model, and b) automatic true orthophoto production. We performed the research in two test areas within the Municipality of Ljubljana. We compared the both procedures and the produced true orthophotos with the classical orthophoto, and assessed the production time component as well. To produce true orthophoto, larger overlapping of aerial images is required, thus the time for aerial surveying is increased by approximately 25 percentage. The time of manual work, compared to classical orthophoto production, is lower by the factor 0.27 for automatically produced true orthophoto, and is greater by the factor 2.75 for the true orthophoto based on digital terrain model and digital building model. Aesthetic appearance of automatically produced orthophoto has only minor shortcomings on the buildings’ roof edges which are slightly serrated. Considering all the mentioned aspects, the automatically produced true orthophoto is competitive with the classical orthophoto.

Neural Network Method for Constructing Three-Dimensional Models of Rigid Objects from Satellite Images

A method has been developed for constructing three-dimensional models of rigid objects on the earth’s surface using one satellite image using the example of railway infrastructure. The method consists in step-by-step processing of satellite images with sequential application of two convolutional neural networks. In the first processing step, a satellite image is segmented by a neural network to select a plurality of objects of predetermined classes. At the second stage of processing with the help of neural network local analysis of image areas detected by results of the first stage of processing is performed. The results of the second processing step are used to estimate the parameters of the 3D model of the object. The possibilities of the method are shown by the example of processing a satellite image of the railway infrastructure. The following classes of informative areas are considered: building, wall edge, roof edge, building shadow, railway infrastructure, car, highway; rails, poles and shadows from poles (taken as reference objects for estimating scaling coefficients in certain directions). An example is given of using the developed method of highlighting typical railway infrastructure objects and for subsequent evaluation of the parameters of a three-dimensional building model partially obscured by trees.

Directional Measurements in Urban Street Canyons From Macro Rooftop Sites at 28 GHz for 90% Outdoor Coverage

Path gain and effective directional gain in azimuth in urban canyons from actual rooftop base station sites are characterized based on a massive data set of 3000 links on 12 streets in two cities, with over 21 million individual continuous wave power measurements at 28 GHz using vertically polarized antennas. Large street-to-street path gain variation is found, with median street path gain varying over 30 dB at similar distances. Coverage in the street directly illuminated by a roof edge antenna is found to suffer an average excess loss of 11 dB relative to free space at 200 m, with empirical slope-intercept fit model representing the data with 7.1 dB standard deviation. Offsetting the base antenna 5 m away from roof edge, as is common in macro cellular deployments, introduces an additional average loss of 15 dB at 100 m, but this additional loss reduces with distance. Around the corner loss is well modeled by a diffraction formula with an empirically obtained diffraction coefficient. Effective azimuthal gain degradation due to scatter is limited to 2 dB for 90% of data, supporting effective use of high gain antennas in urban street canyons.

Salt tectonics and controls on halokinetic-sequence development of an exposed deepwater diapir: The Bakio Diapir, Basque-Cantabrian Basin, Pyrenees

This work evaluates growth strata adjacent to the Bakio Diapir in the Basque Pyrenees, aiming to discuss the application of halokinetic-sequence concepts, mainly developed in shallow-water to subaerial environments, to deepwater depositional settings. This is one of the few exposed passive diapirs developed in deepwater environments, with both synkinematic carbonate and siliciclastic strata. Thus, considering that large hydrocarbon producing areas are located in salt provinces developed in deepwater environments, this study is of interest not only to researchers in salt tectonics, but also to industry geoscientists focusing on hydrocarbon exploration and production. We present a 3D analysis of this outstanding salt structure by integrating detailed geological cartography, high-resolution bathymetry, seismic, and well data. The resulting reconstruction enables us to trace the extent of the diapir offshore and decipher its evolution from its formation as a salt wall developed above the overlap of two basement-involved faults until its squeezing during the Pyrenean compression. But more significantly, it allows us to discern the roles played by bathymetry, subsidence, and sedimentation type in the configuration of halokinetic sequences developed in deepwater environments. Thus, our study shows that: 1) the geometry of halokinetic sequences is defined by the thickness of the roof edges, the dip of the salt-sediment interface in the limbs of the active drape fold, and the onlap angle of the synkinematic sediments over the salt; 2) the roof thickness is controlled not only by the ratio between salt-rise and regional sediment-accumulation rate, but also by the water depth of the diapir roof and the depositional environment that can promote vertical aggradation of a supra-diapir carbonate buildup; and 3) high surface slopes and consequent debrites are not exclusive to, and characteristic of, hook halokinetic sequences and tabular composite halokinetic sequences.

Integrity Assessment of Obsolete Thermal Storage Tank in District Heating System

The thermal storage tank in a district heating system is a component that stores excess heat during normal operation and releases the stored heat to increase the efficiency of the system, when the heat source is stopped or additional demands occur. Recently, an obsolete thermal storage tank was dismantled for the first time since it began operation 30 years ago. In this work, the corrosion integrity of the obsolete thermal storage tank was evaluated by examining its appearance, thickness thinning, corrosion products, microstructure, and mechanical properties. Samples were taken at various locations (roof, shell, bottom) of the thermal storage tank, which enabled diagnosis of the respective environmental degradations. Severe corrosion was found in the roof edge plate due to corrosion under the insulation, and exhibited thinning exceeding ~49% of the designed thickness. In this location, the ferrite-pearlite band structure disappeared and deteriorated microstructures, such as decarburization and spheroidized pearlite, were measured, which resulted in a ~27% decrease in hardness. The inner surfaces of the bottom and shell plate were well covered with a magnetite film, and the degradation of the microstructure and mechanical properties showed a permissible limit in terms of ASTM A285/A516. In addition, no particular drop in hardness was found in the weld zone of each plate.

HOLISTIC PARAMETRIC RECONSTRUCTION OF BUILDING MODELS FROM POINT CLOUDS

Abstract. Building models are conventionally reconstructed by building roof points via planar segmentation and then using a topology graph to group the planes together. Roof edges and vertices are then mathematically represented by intersecting segmented planes. Technically, such solution is based on sequential local fitting, i.e., the entire data of one building are not simultaneously participating in determining the building model. As a consequence, the solution is lack of topological integrity and geometric rigor. Fundamentally different from this traditional approach, we propose a holistic parametric reconstruction method which means taking into consideration the entire point clouds of one building simultaneously. In our work, building models are reconstructed from predefined parametric (roof) primitives. We first use a well-designed deep neural network to segment and identify primitives in the given building point clouds. A holistic optimization strategy is then introduced to simultaneously determine the parameters of a segmented primitive. In the last step, the optimal parameters are used to generate a watertight building model in CityGML format. The airborne LiDAR dataset RoofN3D with predefined roof types is used for our test. It is shown that PointNet++ applied to the entire dataset can achieve an accuracy of 83% for primitive classification. For a subset of 910 buildings in RoofN3D, the holistic approach is then used to determine the parameters of primitives and reconstruct the buildings. The achieved overall quality of reconstruction is 0.08 meters for point-surface-distance or 0.7 times RMSE of the input LiDAR points. This study demonstrates the efficiency and capability of the proposed approach and its potential to handle large scale urban point clouds.

Dynamic response of a large vertical tank impacted by blast fragments from chemical equipment

Target equipment damage is a key factor in the quantitative risk assessment of domino effects caused by blast fragments. However, the damage process of target equipment caused by blast fragments in domino effect accidents has not been studied clearly. In this paper, a finite element (FE) model of large vertical storage tanks impacted by blast fragments is built to study the damage process. A small-scaled fragment impact experiment is carried out to validate the FE modeling approach. Based on accident statistical data and empirical formulas, the reasonable ranges of blast fragment parameters are obtained. The comparison between the rigid and the elastic–plastic fragment impact is conducted. The effects of filling degree, tank volume, impact height, roof-impact angle, fragment type, impact velocity and impact angle on the damage process are discussed. It can be concluded that: (1) The “added mass” effect of the inner liquid on the impact process increases with the filling degree. The fragment deformation has important effects on the impact process. (2) The impact process includes multiple collisions rather than only one collision. (3) The reduction of tank volume and the increase of impact height on tank wall can increase the damage degree. The impact on roof edge can make the fragment penetration easier. (4) End-caps for identical kinetic energy are more harmful to tanks than plates. The end-cap impact with horizontal impact angle of 35° causes the most serious deformation. This study provides insights into the damage mechanism and prevention of domino effects caused by blast fragments.

Unmanned aerial vehicles can accurately, reliably, and economically compete with terrestrial mapping methods

Structure from motion (SfM) and imagery-derived point clouds (IDPC) are excellent tools for collecting spatial data. However, reported accuracies from unmanned aerial systems (UAS) commonly fall short of their theoretical potential. The research presented here, using a DJI Inspire 2 with post-processed kinematic direct geopositioning, demonstrates that UAS mapping can be consistently accurate enough for use in place of, or in concert with, terrestrial methods (2 cm vertical root mean squared error). We further demonstrate that features that are missing or distorted in IDPC (e.g., roof edges, break lines, and above-ground utilities) can be collected from UAS-imagery stereo models with similar accuracy. Accuracy in the experiments was verified by comparison to data from a total station and terrestrial laser scanner (TLS). Use of the recommended hardware and stereo compilation reduced mapping costs by 40%–75% on three test projects.

Correlation of internal and external pressures and net pressure factors for cladding design

Net pressures on roofs and walls of buildings are dependent on the internal and external pressure fluctuations. The variation of internal and external pressures are influenced by the size and location of the openings. The correlation of external and internal pressure influences the net pressures acting on cladding on different parts of the roof and walls. The peak internal and peak external pressures do not occur simultaneously, therefore, a reduction can be applied to the peak internal and external pressures to obtain a peak net pressure for cladding design. A 1:200 scale wind tunnel model study was conducted to determine the correlations of external and internal pressures and effective reduction to net pressures (i.e., net pressure factors, ????????) for roof and wall cladding. The results show that external and internal pressures on the windward roof and wall edges are well correlated. The largest ????????,????????????, highest correlation coefficient and the highest ???????? are obtained for different wind directions within 90o ≤ ???? ≤ 135o , where the large openings are on the windward wall. The study also gives net pressure factors ???????? for areas on the roof and wall cladding for nominally sealed buildings and the buildings with a large windward wall opening. These factors indicate that a 5% to 10% reduction to the action combination factor, ???????? specified in AS/NZS 1170.2(2011) is possible for some critical design scenarios.

OEBR-GAN: Object Extraction and Background Recovery Generative Adversarial Networks

Generative adversarial networks (GAN) have been widely used in the field of image-to-image translation. In this paper, we have proposed a novel object extraction and background recovery (OEBR-GAN) model, which can extract objects from an image and then complete the image by inpainting the background of the image. This model has been developed for a solar panel installation project, where the user would like to input an original colored image of the roof, and as output, the user requires an edge detected roof image. However, the condition in user requirement is that any object that is hiding the roof edges should be removed first and the background of that part of the roof image should be recovered so that the user can obtain a complete connected edge detected image of the roof. Therefore, the model also completes the image by connecting the hidden edges of the roof. We could achieve the user objective by building a GAN model with a dual generator and dual discriminator network. The generators have been built using an encoder-decoder network with and without skip connections and the discriminators have been built using deep convolutional neural networks and encoder architecture. Quantitative comparisons in the result section shows that OEBR-GAN performs much better than other adversarial models on our collected dataset.

AUTOMATIC RECONSTRUCTION OF ROOF MODELS FROM BUILDING OUTLINES AND AERIAL IMAGE DATA

The knowledge of roof shapes is essential for the creation of 3D building models. Many experts and researchers use 3D building models for specialized tasks, such as creating noise maps, estimating the solar potential of roof structures, and planning new wireless infrastructures. Our aim is to introduce a technique for automating the creation of topologically correct roof building models using outlines and aerial image data. In this study, we used building footprints and vertical aerial survey photographs. Aerial survey photographs enabled us to produce an orthophoto and a digital surface model of the analysed area. The developed technique made it possible to detect roof edges from the orthophoto and to categorize the edges using spatial relationships and height information derived from the digital surface model. This method allows buildings with complicated shapes to be decomposed into simple parts that can be processed separately. In our study, a roof type and model were determined for each building part and tested with multiple datasets with different levels of quality. Excellent results were achieved for simple and medium complex roofs. Results for very complex roofs were unsatisfactory. For such structures, we propose using multitemporal images because these can lead to significant improvements and a better roof edge detection. The method used in this study was shared with the Czech national mapping agency and could be used for the creation of new 3D modelling products in the near future.

Empirical and Numerical Analysis of Aerodynamic Drag on a Typical SUV Car Model at Different Locations of Vortex Generator

The aerodynamic characteristics are concerned with the fuel consumption rate and the stability of a high speed vehicle. The current research aims at studying the aerodynamic behavior of a typical SUV vehicle model mounted with the vortex generator (VG) at various linear positions with reference to its rear roof edge. The flow field around the vehicle model was observed at different wind speed conditions. It had been determined that at the instance of lower wind speed, the VG had minimal effects of aerodynamic drag on the vehicle body. However, at the instance of higher wind speed conditions the magnitude of the drag force decreased significantly. Vehicles move at higher speeds in the highways, location of the VG varied towards the upstream of the vehicle due to early flow separation. Therefore test were conducted at different wind speeds and locations of VG. The numerical simulation conduced in this study provides flow characteristics around the vehicle model for different wind speeds. The realizable k−ε model was used to simulate and validate the empirical results in an effective manner. By using experimental data, the drag was reduced by 9.04 % at the optimized VG location. The results revealed that the induced aerodynamic drag would determine the best car shape. This paper provides a better understanding of VG positioning for enhanced flow separation control.

Research on the Stress Variation Effect of Overburden Loads in Utility Tunnel Roof

As an underground structure, the utility tunnel is subjected to the pressure of overburden soil. Calculation of the overburden pressure soil are different when the burial depth changes. In this paper, the calculation methods of soil loads are summarized. In view of utility tunnel, the ANSYS finite element analysis is adopted to analyze the pressure of the roof. When calculating the structural stress, the working condition is considered as the backfill after the excavation of the soil. Considering the stress variation of the roof edge of the structure with different burial depth, it can be found that compressive stress at 1/2 span is about 1.7-2.2 times of the tensile stress at the corner. With the increase of burial depth, the rate of increasing roof stress value decreases. When the burial depth reaches a certain depth, soil arching effect plays a role and the roof stress value tends to be stable. Some suggestions are given to the designers of the utility tunnel which can reduce the waste of building materials.

A new regression model to predict BIPV cell temperature for various climates using a high-resolution CFD microclimate model

ABSTRACT Understanding of cell temperature of Building Integrated Photovoltaics (BIPV) is essential in the calculation of their conversion efficiency, durability and installation costs. Current PV cell temperature models mainly fail to provide accurate predictions in complex arrangement of BIPVs under various climatic conditions. To address this limitation, this paper proposes a new regression model for prediction of the BIPV cell temperature in various climates and design conditions, including the effects of relative PV position to the roof edge, solar radiation intensity, wind speed, and wind direction. To represent the large number of possible climatic and design scenarios, the advanced technique of Latin Hypercube Sampling was firstly utilized to reduce the number of investigated scenarios from 13,338 to 374. Then, a high-resolution validated full-scale 3-dimensional Computational Fluid Dynamics (CFD) microclimate model was developed for modelling of BIPV’s cell temperature, and then was applied to model all the reduced scenarios. A nonlinear multivariable regression model was afterward fit to this population of 374 sets of CFD simulations. Eventually, the developed regression model was evaluated with new sets of unused climatic and design data when a high agreement with a mean discrepancy of 3% between the predicted and simulated BIPV cell temperatures was observed.

Numerical Investigation of Wind Pressure Coefficients for Photovoltaic Arrays Mounted on Building Roofs

The wind pressure distribution on the photovoltaic (PV) array is of great importance to the wind resistance design. The flow field related to the pressure can be influenced significantly by the turbulence induced by the building roof edge (Kopp et al., 2012) and it is essential to consider the building effect during the investigation. However, most CFD (computational fluid dynamics) investigations of wind pressure distribution on the PV array are limited to ground mounted PV array without the building. There is a necessity to extend the application of CFD method to flows around roof-mounted PV array. This study investigated the wind pressure distributions on PV arrays mounted on building roofs by means of Reynolds-averaged Navier-Stokes (RANS) approach using the FLUENT software. Since RANS models are sensitive to the fiow condition, several RANS models are adopted and the most accurate RANS model for predicting this type of flow is identified based on a comparison with the wind tunnel experimental results. The SST Κ-ω model can predict the highest net mean wind uplift located at panels upstream accurately and is further applied for the parameter analysis to facilitate the installation of roof-mounted PV array. Numerical simulations of the wind flow field for wind angles between 0° to 180° were carried out at intervals of 20°, and the resulted net pressure distributions were presented. The influences of tilt angles of the PV array were investigated and the pressure distributions of PV panels were related to the flow field. Moreover, the effects of clearance between the PV array and building roof on the flow fields and pressure distributions of the PV array related to PV array tilt angle are studied.