L. Díaz-Vilariño

3D Modeling of Building Indoor Spaces and Closed Doors from Imagery and Point Clouds

3D models of indoor environments are increasingly gaining importance due to the wide range of applications to which they can be subjected: from redesign and visualization to monitoring and simulation. These models usually exist only for newly constructed buildings; therefore, the development of automatic approaches for reconstructing 3D indoors from imagery and/or point clouds can make the process easier, faster and cheaper. Among the constructive elements defining a building interior, doors are very common elements and their detection can be very useful either for knowing the environment structure, to perform an efficient navigation or to plan appropriate evacuation routes. The fact that doors are topologically connected to walls by being coplanar, together with the unavoidable presence of clutter and occlusions indoors, increases the inherent complexity of the automation of the recognition process. In this work, we present a pipeline of techniques used for the reconstruction and interpretation of building interiors based on point clouds and images. The methodology analyses the visibility problem of indoor environments and goes in depth with door candidate detection. The presented approach is tested in real data sets showing its potential with a high door detection rate and applicability for robust and efficient envelope reconstruction.

Automatic Segmentation and Shape-Based Classification of Retro-Reflective Traffic Signs from Mobile LiDAR Data

Recently, many studies have demonstrated the valid contribution of mobile laser scanning to road safety improvements, thus intense efforts have been made to implement automatic data processing using laser scanning data, with special emphasis on road object recognition. This study is focused on the detection and classification of retro-reflective vertical traffic signs according to their function (danger, give way, prohibition/obligation, and indication) from mobile laser scanning data by considering geometric and radiometric information, but without relying on trajectory data. The global strategy for segmentation involves the application of an optimized intensity threshold in order to segment the points that correspond to traffic sign panels. Next, contour recognition is performed for each sign using a linear regression model based on a raster image, which is generated for each cluster of points. The shape evaluation is motivated by the correspondence between contour shape and function of the traffic. The completeness of results for detection (92.11%) and classification (83.91%) demonstrates that this implementation is promising for the automatic detection and inventory analysis of traffic signs in road mapping applications. The efficiency rates are acceptable in urban areas, but our tests indicate that the detection and classification rates are more robust in road environments.

Automatic Detection and Segmentation of Columns in As-Built Buildings from Point Clouds

Over the past few years, there has been an increasing need for tools that automate the processing of as-built 3D laser scanner data. Given that a fast and active dimensional analysis of constructive components is essential for construction monitoring, this paper is particularly focused on the detection and segmentation of columns in building interiors from incomplete point clouds acquired with a Terrestrial Laser Scanner. The methodology addresses two types of columns: round cross-section and rectangular cross-section. Considering columns as vertical elements, the global strategy for segmentation involves the rasterization of a point cloud onto the XY plane and the implementation of a model-driven approach based on the Hough Transform. The methodology is tested in two real case studies, and experiments are carried out under different levels of data completeness. The results show the robustness of the methodology to the presence of clutter and partial occlusion, typical in building indoors, even though false positives can be obtained if other elements with the same shape and size as columns are present in the raster.

As-built BIM with shades modeling for energy analysis

The use of Building Information Models (BIM) for energy analysis is becoming a common application, supported by the appearance of standards and regulations restricting energy consumption and energy efficiency in the building sector. BIMs from already built buildings are being generated with the help of high-technology devices such as laser scanners, which acquire the physical reality of a scene with high accuracy in a short time. However, the environment of the building, and especially surfaces producing shades, which are essential for the performance of meaningful energy studies, is usually forgotten as the focus is set on the representation of complex geometries. With the aim of generating a BIM able to be subjected to energy analysis, this paper presents a working methodology including data acquisition with a laser scanner, shape extraction of the building itself and its surroundings, and conversion of extracted elements, including shade surfaces, to BIM components.

Quantitative Evaluation of CHT and GHT for Column Detection under Different Conditions of Data Quality

AbstractQuality control and project monitoring are topics of interest in the field of architectural-engineering-construction/facility management. The need to automatise the process and analyze data...

Interurban visibility diagnosis from point clouds

Abstract We present an approach for automatic visibility analysis in interurban roads from point clouds. The methodology is based on a ray-tracing algorithm followed by an occlusion detection to identify potential obstacles between the driver and the theoretical position of pedestrians and cyclists. As a result, the area of visibility from each driver position is obtained. The method compares the performance and suitability of point clouds acquired from both Airborne and Mobile Laser Scanning. The methodology is tested in six real case studies. In most cases, results obtained from MLS are more accurate since the point clouds are acquired from a perspective similar to driver and they have higher resolution.

3D reconstruction of cubic armoured rubble mound breakwaters from incomplete lidar data

This manuscript proposes an algorithm for the reconstruction of the cube armoured rubble mound breakwaters’ geometry from incomplete lidar point clouds in order to perform structural monitoring over time. Rubble mound breakwaters are critical structures used in the protection of beaches and ports. The constant wave actions result in the degradation of these defences, causing catastrophic coastal damage. Early detection of the degradation of the breakwaters is a key topic to prevent disasters. Remote-sensing techniques such as lidar and photogrammetry contribute to the monitoring of civil engineering structures. The reconstruction of cube armoured breakwaters relies on normal-vector segmentation and a priori cube’s properties. For the successful modelling of the armour units, two options are available, the first one requires three perpendicular planes of the cube to be scanned and segmented, while for the second, only two planes are required. The resulting models were compared to ground truth handmade delineation of 1.25 m side length cubed armour units, leading to an accuracy range of 7–15 cm. Besides that, the precision results are constrained to 2.7–9.9 mm for the three plane-based reconstruction and 7.2–14.5 mm for the two plane-based one. The reconstructed armour units are well defined and suitable to be used as a 3D computer-aided design (CAD) model for monitoring breakwaters.

Autonomous Point Cloud Acquisition of Unknown Indoor Scenes

This paper presents a methodology for the automatic selection of heuristic scanning positions in unknown indoor environments. The surveying is carried out by a robotic system following a stop-and-go procedure. Starting with a random scan position in the room, the point cloud is discretized in voxels and they are submitted to a two-step classification and are labelled as occupied, occluded, empty, window, door, or exterior based on a visibility analysis. The main objective of the methodology is to obtain a complete point cloud of the indoor space and accordingly, the next best position is the scan position minimizing occluded voxels. Because the method locates doors and windows, the room can be delimited and the scan can continue for adjacent rooms. This approach has been tested in a real case study, in which three scans were developed.

Wave Run-Up Monitoring on Rubble-Mound Breakwaters Using a Photogrammetric Methodology

AbstractWave overtopping in rubble-mound breakwaters is an aspect that must be controlled in order to ensure the correct operation of port facilities. This manuscript shows a photogrammetric methodology to automatically monitor this process based on the use of low-cost cameras. Cameras are used for acquiring video of breakwaters. Then, image processing algorithms based on binarization tresholding and projective transformation, combined with geometric data of the structure provided by the facility manager, are used to evaluate the run-up height and automatically detect the existence or absence of a overtopping event. The methodology was successfully tested in the outer port of Punta Langosteira, A Coruna, Spain.

Influence of mobile light detecting and ranging data quality in road runoff evaluation

Abstract. A mobile light detecting and ranging (LiDAR) system is used to provide point cloud datasets as a topographic base for runoff studies. The point clouds are rasterized to evaluate road runoff using the D8 algorithm. Gaussian noise is artificially induced in the point cloud to simulate inaccuracies in geopositioning and determine its influence in the evaluation of runoff direction. Accuracy in the determination of flow direction decreases with the increase of Gaussian noise. Accuracy also decreases with the decrease of the cell size of the raster dataset. Flow direction shows inaccuracies up to 47 deg with a cell resolution of 0.5 m and Gaussian noise of 0.15 m (standard deviation). On the other hand, cell resolutions of 5 m show a maximum difference of 15 deg with the same noise.