Point Clouds at the service of BIM

A point cloud is the result of the detection and ranging of a certain object by laser scanner. Learn how it simplifies BIM modeling

Every building intervention on an existing building must start from the reproduction of the initial situation, the so-called “Survey Stage”. Today, thanks to increasingly efficient technologies that allow us to optimize the entire process of surveying and reproducing the building by means of 3D models, it is now possible to create a profitable synergy between the design activity operated in BIM (Building Information Modeling) environment and the necessary preventive phase of surveying the starting condition of the artefact.

Scan to BIM is a reverse engineering process that uses advanced sensing technologies, such as 3D laser scanning, Structure for Motion and high-definition photogrammetry, to obtain point clouds and 3D meshes useful as the basis for the BIM modeling process.

The result of the laser scanner survey returns a set of millions of points in a 3-dimensional space; the obtained cluster of points (point cloud) can be managed with a point cloud to BIM software in order to produce a digital model of the existing one, which can also be evaluated in a virtual reality environment.

Edificius interface with the point cloud import function

What is a Point Cloud

A point cloud is the result of the ranging and detection of a certain object by laser scanner. In easy terms, it’s a cluster of many points arranged in space containing information such as position (x,y,z coordinates), color, reflectance, etc.

The point cloud can also be obtained through structure from motion techniques that allow you to obtain points by processing appropriately taken photos; it can also be displayed through special software that allows you to manage and graphically return the data obtained from the survey.

But, once you get the point cloud (with one of the techniques available today), what are its advantages?

There are many benefits beyond just visualization. In fact you can:

• do comparisons between different models
• use the cloud to create the BIM model of the building
• measure distances directly on the cloud model
• generate a mesh model (surface) obtained by joining various nodes in space.

This depends on the purpose of the survey and the quality of the point cloud obtained by the survey. It is generally referred to as cloud resolution, linked to the density of the points.

We have said that with the point cloud it is also possible to obtain a 3D model composed of mesh: each mesh is formed by many triangles described by three-dimensional x,y,z coordinates present in the point cloud.

The mesh originally obtained has no color. The color (or texture) can be assigned to the various polygons in two different ways:

• color for vertex – the “average” of the colors of the various points that form the vertices of the polygon is taken into account. In this case, the quality of the final result is directly linked to the resolution of the model, since it depends on the points that form the model itself.
• texture – the photos taken by the survey instrument are taken as a reference. These images are already oriented with respect to the model and can be projected directly onto the surfaces. Each mesh will be associated with the union of these photos (textures) that can be seen as a “sheet” that covers the polygons. The quality of the final result will depend on the quality of the photographs and no longer on the resolution of the model.

What are the survey techniques used

There are different techniques used for surveying buildings. Without dwelling on the traditional ones (for example trilateration), we see those that allow the point cloud to be generated.

Photogrammetry

Photogrammetry is a survey technique that allows you to obtain position, shape and size of the object to be detected by processing a series of images. This is possible thanks to the analysis of stereoscopic frames, obtained thanks to a stereoscopic camera or with a metric camera that frames the same object and that is arranged in succession in two distinct positions.

The photogrammetric process can be divided into 3 phases:

• the socket: photograph the object with a stereometric camera or with the camera placed in two different points;
• frame orientation: the stereometric frames are suitably oriented in order to obtain a three-dimensional model;
• reception: the 3D model is reconstructed and on it it is possible to detect shape and size.

Illustration of the photogrammetric survey working principle

The photogrammetric survey can be aerial, using devices such as drones, or terrestrial, using any camera or stereoscopic camera.

Drones

A drone is a remotely piloted aircraft (RPA), meaning that it does not provide for any human presence on board the aircraft for flying, but a specific remote control is used.

There are many types of drones available on the market and that can be grouped into these three macrocategories:

• propeller-structured drones;
• drones with planar structure;
• hybrid drones.

These tools allow you to produce aero-photogrammetric surveys ensuring a faithful reproduction of the object. Here are the basic operations to follow:

• identification of the area to be detected (for example through Google Maps)
• area analysis and intake design
• selection of the take-off and landing points of the drone and identification of the positioning point of the ground control station
• flight plan operations by checking for any obstructions
• planning of any automatic flight missions
• identification of ground measurements that can help and support photogrammetric survey
• image acquisition
• frame orientation and Scene Reconstruction (Structure for Motion)
• point cloud generation
• generation of the 3D mesh model.

To do this you can rely on two flight mission methodologies:

• automatic mode using WayPoints, which automatically scans an area by selecting points from the reference map planned in advance;
• manual mode, takes photos according to a predetermined pattern.

Cloud of points in BIM | Drone used in a nadiral survey methodology

Laser scanners

LASER scanners detect and range objects thanks to a laser beam. The laser light hits the object surface and in relation to the bounce and return time (time necessary for the beam to return to the emitter device), the angle of the device and other factors, it can identify the exact position in space of each single point of the object.

Laser scanners can be divided into three categories:

• time of Flight (TOF) – is the distance measured by identifying the istantaneous time difference between the light emission and its return back to the device receptors. In this case the device that emits the signal also functions as a receiver (Ranging scanner).
• unlike phase shift – the distance is measured knowing the phase difference between the phase of the laser wave at the time the signal is emitted and the subsequent reception of the return pulse. Even in this situation the device that emits the signal is the same being received (Ranging scanner).
• triangulation – the measurement is calculated thanks to the triangulation technique with resulting trigonometric calculations. In this case, the device that emits the signal and the one that receives it, are separated and placed at a constant and calibrated distance. This system allows us obtain a much more detailed survey but at closer distances.

The final result of the survey is a cloud of scaled points that reflect the detected object. This model consists of millions of points with their spatial coordinates and knowledge of their material reflectance values. In addition to that, thanks to integrated cameras, each object can also be mapped in colour with a precise chromatic detection process for each point. This improves the overall quality of the survey returning excellent imagery of scanned objects.

The limits of the laser scanner survey system are mainly determined by the maximum distance of the scanning equiment from the object, any non-visible surfaces and shaded areas. These are all problems that are solved thanks to the survey process from different points allowing surveyors to obtain a much more faithful model with smaller error margins.

Example image of the laser scanner survey – point cloud in BIM

Laser scanners with SLAM TECHNOLOGY

Among the most innovative systems we also have the SLAM laser scanner (Simultaneous Localization And Mapping). This type of equipment allows you to perform scans while in motion and without any need to identify fixed points for the survey.

With this type of laser scanner technology, you can move around in space and recreate a map of the environment by automatically locating yourself inside it (therefore without using any GPS support).

GeoSLAM, mobile system with SLAM technology. Source: Microgeo.it

These devices can be installed on vehicles and drones, so that they can also survey very large surfaces in a relatively short time, or even simply transported by hand within environments where the GPS signal is not present.

These scanners are classified into the iMMS subcategory (indoor Mobile Mapping Systems) belonging to the largest group of MMS (Mobile Mapping Systems). Although the former have been optimized for indoor survey, they are also often used outdoors to scan infrastructure, construction sites, buildings, etc.

Both systems (MMS and iMMS) should not be confused with 3D LiDAR (Light Detection and Ranging) scanners as in this case a broader category is referred to. What you need to know, however, is that LiDAR technology (technology that uses laser light to make measurements) is used for almost all MMS and iMMS systems.

Point cloud in BIM | Survey with slam technology

Point Cloud Management and BIM Model Generation

The point cloud, as mentioned, is fundamental to the process of generating the parametric model.

There are point cloud BIM software, such as ACCA’s Edificius software, which allows you to take advantage of powerful recognition tools to automatically identify the different elements present in the model.

Creation of the BIM model from a point cloud

You can therefore detect objects such as walls, doors, windows, floors, in an automatic and precise way, obtaining a digital model capable of automatically returning the necessary models and drawings:

• construction drawings (plans, elevations, sections, etc.);
• technical reports;
• integrated bill of quantities;
• activity scheduling;
• etc.

Experience how easy it is to build up a BIM model from a point cloud for yourself. Download the trial version of this Point Cloud to BIM software and import your point cloud. Otherwise, simply download one of these sample files present in the usBIM – BIM Management System.