There are a variety of technologies for digitally acquiring the shape of a 3D object. The techniques work with most or all sensor types including optical, acoustic, laser scanning, radar, thermal, and seismic. 3D scan technologies can be split in 2 categories: contact and non-contact. Non-contact solutions can be further divided into two main categories, active and passive. There are a variety of technologies that fall under each of these categories.
Contact (CMM) with scanning head Contact 3D scanners work by physically probing (touching) the part and recording the position of the sensor as the probe moves around the part. There are two main types of contact 3D scanners: •
Coordinate measuring machines (CMMs) which traditionally have 3 perpendicular moving axis with a touch probe mounted on the Z axis. As the touch probe moves around the part, sensors on each axis record the position to generate XYZ coordinates. Modern CMMs are 5 axis systems, with the two extra axes provided by pivoting sensor heads. CMMs are the most accurate form of 3D measurement achieving micron precision. The greatest advantage of a CMM after accuracy is that it can be run in autonomous (CNC) mode or as a manual probing system. The disadvantages of CMMs are their upfront cost and the technical knowledge required to operate them. •
Articulated Arms which generally have multiple segments with polar sensors on each joint. As per the CMM, as the articulated arm moves around the part sensors record their position and the location of the end of the arm is calculated using complex math and the wrist rotation angle and hinge angle of each joint. While not usually as accurate as CMMs, articulated arms still achieve high accuracy and are cheaper and slightly easier to use. They do not usually have CNC options. Both modern CMMs and Articulated Arms can also be fitted with non-contact laser scanners instead of touch probes.
Non-contact active Active scanners emit some kind of radiation or light and detect its reflection or radiation passing through object in order to probe an object or environment. Possible types of emissions used include light,
ultrasound or
x-ray.
Time-of-flight scanner may be used to scan buildings, rock formations, etc., to produce a 3D model. The lidar can aim its laser beam in a wide range: its head rotates horizontally, a mirror flips vertically. The laser beam is used to measure the distance to the first object on its path. The time-of-flight 3D laser scanner is an active scanner that uses laser light to probe the subject. At the heart of this type of scanner is a time-of-flight
laser range finder. The laser range finder finds the distance of a surface by timing the round-trip time of a pulse of light. A laser is used to emit a pulse of light and the amount of time before the reflected light is seen by a detector is measured. Since the
speed of light c is known, the round-trip time determines the travel distance of the light, which is twice the distance between the scanner and the surface. If t is the round-trip time, then distance is equal to \textstyle c \! \cdot \! t / 2. The accuracy of a time-of-flight 3D laser scanner depends on how precisely we can measure the t time: 3.3
picoseconds (approx.) is the time taken for light to travel 1 millimetre. The laser range finder only detects the distance of one point in its direction of view. Thus, the scanner scans its entire field of view one point at a time by changing the range finder's direction of view to scan different points. The view direction of the laser range finder can be changed either by rotating the range finder itself, or by using a system of rotating mirrors. The latter method is commonly used because mirrors are much lighter and can thus be rotated much faster and with greater accuracy. Typical time-of-flight 3D laser scanners can measure the distance of 10,000~100,000 points every second. Time-of-flight devices are also available in a 2D configuration. This is referred to as a
time-of-flight camera.
Triangulation Triangulation based 3D laser scanners are also active scanners that use laser light to probe the environment. With respect to time-of-flight 3D laser scanner the triangulation laser shines a laser on the subject and exploits a camera to look for the location of the laser dot. Depending on how far away the laser strikes a surface, the laser dot appears at different places in the camera's field of view. This technique is called triangulation because the laser dot, the camera and the laser emitter form a triangle. The length of one side of the triangle, the distance between the camera and the laser emitter is known. The angle of the laser emitter corner is also known. The angle of the camera corner can be determined by looking at the location of the laser dot in the camera's field of view. These three pieces of information fully determine the shape and size of the triangle and give the location of the laser dot corner of the triangle. In most cases a laser stripe, instead of a single laser dot, is swept across the object to speed up the acquisition process. The use of
triangulation to measure distances dates to antiquity.
Strengths and weaknesses Time-of-flight range finders are capable of operating over long distances on the order of kilometres. These scanners are thus suitable for scanning large structures like buildings or geographic features. A disadvantage is that, due to the high speed of light, measuring the round-trip time is difficult and so the accuracy of the distance measurement is relatively low, on the order of millimetres. Triangulation range finders, on the other hand, have a range of usually limited to a few meters for reasonably sized devices, but their accuracy is relatively high. The accuracy of triangulation range finders is on the order of tens of
micrometers. Time-of-flight scanners' accuracy can be lost when the laser hits the edge of an object because the information that is sent back to the scanner is from two different locations for one laser pulse. The coordinate relative to the scanner's position for a point that has hit the edge of an object will be calculated based on an average and therefore will put the point in the wrong place. When using a high resolution scan on an object the chances of the beam hitting an edge are increased and the resulting data will show noise just behind the edges of the object. Scanners with a smaller beam width will help to solve this problem but will be limited by range as the beam width will increase over distance. Software can also help by determining that the first object to be hit by the laser beam should cancel out the second. At a rate of 10,000 sample points per second, low resolution scans can take less than a second, but high resolution scans, requiring millions of samples, can take minutes for some time-of-flight scanners. The problem this creates is distortion from motion. Since each point is sampled at a different time, any motion in the subject or the scanner will distort the collected data. Thus, it is usually necessary to mount both the subject and the scanner on stable platforms and minimise vibration. Using these scanners to scan objects in motion is very difficult. Recently, there has been research on compensating for distortion from small amounts of vibration and distortions due to motion and/or rotation. Short-range laser scanners can not usually encompass a
depth of field more than 1 meter. When scanning in one position for any length of time slight movement can occur in the scanner position due to changes in temperature. If the scanner is set on a tripod and there is strong sunlight on one side of the scanner then that side of the tripod will expand and slowly distort the scan data from one side to another. Some laser scanners have level compensators built into them to counteract any movement of the scanner during the scan process.
Conoscopic holography In a
conoscopic system, a laser beam is projected onto the surface and then the immediate reflection along the same ray-path are put through a conoscopic crystal and projected onto a CCD. The result is a
diffraction pattern, that can be
frequency analyzed to determine the distance to the measured surface. The main advantage with conoscopic holography is that only a single ray-path is needed for measuring, thus giving an opportunity to measure for instance the depth of a finely drilled hole.
Hand-held laser scanners Hand-held laser scanners create a 3D image through the triangulation mechanism described above: a laser dot or line is projected onto an object from a hand-held device and a sensor (typically a
charge-coupled device or
position sensitive device) measures the distance to the surface. Data is collected in relation to an internal coordinate system and therefore to collect data where the scanner is in motion the position of the scanner must be determined. The position can be determined by the scanner using reference features on the surface being scanned (typically adhesive reflective tabs, but natural features have been also used in research work) or by using an external tracking method. External tracking often takes the form of a
laser tracker (to provide the sensor position) with integrated camera (to determine the orientation of the scanner) or a
photogrammetric solution using 3 or more cameras providing the complete
six degrees of freedom of the scanner. Both techniques tend to use
infrared light-emitting diodes attached to the scanner which are seen by the camera(s) through filters providing resilience to ambient lighting. Data is collected by a computer and recorded as data points within
three-dimensional space, with processing this can be converted into a triangulated mesh and then a
computer-aided design model, often as
non-uniform rational B-spline surfaces. Hand-held laser scanners can combine this data with passive, visible-light sensors — which capture surface textures and colors — to build (or "
reverse engineer") a full 3D model.
Structured light Structured-light 3D scanners project a pattern of light on the subject and look at the deformation of the pattern on the subject. The pattern is projected onto the subject using either an
LCD projector or other stable light source. A camera, offset slightly from the pattern projector, looks at the shape of the pattern and calculates the distance of every point in the field of view. Structured-light scanning is still a very active area of research with many research papers published each year. Perfect maps have also been proven useful as structured light patterns that solve the
correspondence problem and allow for error detection and error correction. The advantage of structured-light 3D scanners is speed and precision. Instead of scanning one point at a time, structured light scanners scan multiple points or the entire field of view at once. Scanning an entire field of view in a fraction of a second reduces or eliminates the problem of distortion from motion. Some existing systems are capable of scanning moving objects in real-time. A real-time scanner using digital fringe projection and phase-shifting technique (certain kinds of structured light methods) was developed, to capture, reconstruct, and render high-density details of dynamically deformable objects (such as facial expressions) at 40 frames per second. Recently, another scanner has been developed. Different patterns can be applied to this system, and the frame rate for capturing and data processing achieves 120 frames per second. It can also scan isolated surfaces, for example two moving hands. By utilising the binary defocusing technique, speed breakthroughs have been made that could reach hundreds to thousands of frames per second.
Modulated light Modulated light 3D scanners shine a continually changing light at the subject. Usually the light source simply cycles its amplitude in a
sinusoidal pattern. A camera detects the reflected light and the amount the pattern is shifted by determines the distance the light travelled. Modulated light also allows the scanner to ignore light from sources other than a laser, so there is no interference.
Volumetric techniques Medical Computed tomography (CT) is a medical imaging method which generates a three-dimensional image of the inside of an object from a large series of two-dimensional X-ray images;
magnetic resonance imaging (MRI) is another similar medical imaging technique, only MRI provides much greater contrast between the different soft tissues of the body than computed tomography (CT) does, making it especially useful in neurological (brain), musculoskeletal, cardiovascular, and oncological (cancer) imaging. These techniques produce a
discrete 3D volumetric representation that can be directly
visualised, manipulated or converted to traditional 3D surface by mean of
isosurface extraction algorithms.
Industrial Although most common in medicine,
industrial computed tomography,
microtomography and MRI are also used in other fields for acquiring a digital representation of an object and its interior, such as non destructive materials testing,
reverse engineering, or studying biological and paleontological specimens.
Non-contact passive Passive 3D imaging solutions do not emit any kind of radiation themselves, but instead rely on detecting reflected ambient radiation. Most solutions of this type detect visible light because it is a readily available ambient radiation. Other types of radiation, such as infrared could also be used. Passive methods can be very cheap, because in most cases they do not need particular hardware but simple digital cameras. •
Stereoscopic systems usually employ two video cameras, slightly apart, looking at the same scene. By analysing the slight differences between the images seen by each camera, it is possible to determine the distance at each point in the images. This method is based on the same principles driving human
stereoscopic vision. •
Photometric systems usually use a single camera, but take multiple images under varying lighting conditions. These techniques attempt to invert the image formation model in order to recover the surface orientation at each pixel. •
Silhouette techniques use outlines created from a sequence of photographs around a three-dimensional object against a well contrasted background. These
silhouettes are extruded and intersected to form the
visual hull approximation of the object. With these approaches some concavities of an object (like the interior of a bowl) cannot be detected.
Photogrammetric non-contact passive methods Photogrammetry provides reliable information about 3D shapes of physical objects based on analysis of photographic images. The resulting 3D data is typically provided as a 3D point cloud, 3D mesh or 3D points. Modern photogrammetry software applications automatically analyze a large number of digital images for 3D reconstruction, however manual interaction may be required if the software cannot automatically determine the 3D positions of the camera in the images which is an essential step in the reconstruction pipeline. Various software packages are available including
PhotoModeler, Geodetic Systems,
Autodesk ReCap,
RealityCapture and
Agisoft Metashape (see
comparison of photogrammetry software). •
Close range photogrammetry typically uses a handheld camera such as a
DSLR with a
fixed focal length lens to capture images of objects for 3D reconstruction. Subjects include smaller objects such as a
building facade, vehicles, sculptures, rocks, and shoes. •
Camera Arrays can be used to generate 3D point clouds or meshes of live objects such as people or pets by synchronizing multiple cameras to photograph a subject from multiple perspectives at the same time for 3D object reconstruction. •
Wide angle photogrammetry can be used to capture the interior of buildings or enclosed spaces using a
wide angle lens camera such as a
360 camera. •
Aerial photogrammetry uses
aerial images acquired by satellite, commercial aircraft or
UAV drone to collect images of buildings, structures and terrain for 3D reconstruction into a point cloud or mesh.
Acquisition from acquired sensor data Semi-automatic building extraction from
lidar data and high-resolution images is also a possibility. Again, this approach allows modelling without physically moving towards the location or object. From airborne lidar data, digital surface model (DSM) can be generated and then the objects higher than the ground are automatically detected from the DSM. Based on general knowledge about buildings, geometric characteristics such as size, height and shape information are then used to separate the buildings from other objects. The extracted building outlines are then simplified using an orthogonal algorithm to obtain better cartographic quality. Watershed analysis can be conducted to extract the ridgelines of building roofs. The ridgelines as well as slope information are used to classify the buildings per type. The buildings are then reconstructed using three parametric building models (flat, gabled, hipped).
Acquisition from on-site sensors Lidar and other terrestrial laser scanning technology offers the fastest, automated way to collect height or distance information. lidar or laser for height measurement of buildings is becoming very promising. Commercial applications of both airborne lidar and ground laser scanning technology have proven to be fast and accurate methods for building height extraction. The building extraction task is needed to determine building locations, ground elevation, orientations, building size, rooftop heights, etc. Most buildings are described to sufficient details in terms of general polyhedra, i.e., their boundaries can be represented by a set of planar surfaces and straight lines. Further processing such as expressing building footprints as polygons is used for data storing in
GIS databases. Using laser scans and images taken from ground level and a bird's-eye perspective, Fruh and Zakhor present an approach to automatically create textured 3D city models. This approach involves registering and merging the detailed facade models with a complementary airborne model. The airborne modeling process generates a half-meter resolution model with a bird's-eye view of the entire area, containing terrain profile and building tops. Ground-based modeling process results in a detailed model of the building facades. Using the DSM obtained from airborne laser scans, they localize the acquisition vehicle and register the ground-based facades to the airborne model by means of Monte Carlo localization (MCL). Finally, the two models are merged with different resolutions to obtain a 3D model. Using an airborne laser altimeter, Haala, Brenner and Anders combined height data with the existing ground plans of buildings. The ground plans of buildings had already been acquired either in analog form by maps and plans or digitally in a 2D GIS. The project was done in order to enable an automatic data capture by the integration of these different types of information. Afterwards virtual reality city models are generated in the project by texture processing, e.g. by mapping of terrestrial images. The project demonstrated the feasibility of rapid acquisition of 3D urban GIS. Ground plans proved are another very important source of information for 3D building reconstruction. Compared to results of automatic procedures, these ground plans proved more reliable since they contain aggregated information which has been made explicit by human interpretation. For this reason, ground plans, can considerably reduce costs in a reconstruction project. An example of existing ground plan data usable in building reconstruction is the
Digital Cadastral map, which provides information on the distribution of property, including the borders of all agricultural areas and the ground plans of existing buildings. Additionally information as street names and the usage of buildings (e.g. garage, residential building, office block, industrial building, church) is provided in the form of text symbols. At the moment the Digital Cadastral map is built up as a database covering an area, mainly composed by digitizing preexisting maps or plans.
Cost • Terrestrial laser scan devices (pulse or phase devices) + processing software generally start at a price of €150,000. Some less precise devices (as the Trimble VX) cost around €75,000. • Terrestrial lidar systems cost around €300,000. • Systems using regular still cameras mounted on RC helicopters (
Photogrammetry) are also possible, and cost around €25,000. Systems that use still cameras with balloons are even cheaper (around €2,500), but require additional manual processing. As the manual processing takes around one month of labor for every day of taking pictures, this is still an expensive solution in the long run. • Obtaining satellite images is also an expensive endeavor. High resolution stereo images (0.5 m resolution) cost around €11,000. Image satellites include Quikbird and Ikonos. High resolution monoscopic images cost around €5,500. Somewhat lower resolution images (e.g. from the CORONA satellite, with a 2 m resolution) cost around €1,000 per 2 images. Note that
Google Earth images are too low in resolution to make an accurate 3D model. == Reconstruction ==