No contact
Active scanners emit some kind of signal and analyze its return to capture the geometry of an object or scene. Electromagnetic radiation (from radio waves to X-rays) or ultrasound are used.
A 3D time-of-flight scanner determines the distance to the scene by timing the round trip of a light pulse. A laser diode emits a pulse of light and the time until the reflected light is seen by a detector is timed. As the speed of light C is known, the round trip time determines the distance of the light's travel, which is twice the distance between the scanner and the surface. If T is the time of the entire trip, then the distance is equal to (C * T)/2. Clearly the certainty of a 3D time-of-flight laser scanner depends on the precision with which the time T can be measured: 3.3 picoseconds (approx.) is the time required for light to travel 1 millimeter. Visible (green) or invisible (near infrared) lasers are used.
The laser distance meter only measures the distance of a point in its scene direction. To carry out the complete measurement, the scanner varies the direction of the distance meter after each measurement, either by moving the distance meter or deflecting the beam using an optical system. This last method is commonly used because the small elements that make it up can be rotated much faster and with greater precision. Typical time-of-flight laser scanners can measure the distance of 10,000 ~ 100,000 points every second.
Features Summary:
Some examples of time-of-flight based scanners:.
The 3D triangulation laser scanner is also an active scanner that uses laser light to examine the environment. The laser light beam hits the object and a camera is used to find the location of the laser spot. Depending on the distance at which the laser hits a surface, the laser dot appears in different places on the camera sensor.
This technique is called triangulation because the laser point, camera and laser emitter form a triangle. The length of one side of the triangle defined by the camera and the laser emitter is known. The vertex angle of the laser emitter is also known. The camera vertex angle (parallax) can be determined by looking at the location of the laser spot on the camera. These three values allow us to determine the rest of the dimensions of the triangle, and therefore, the position of each point in space.
The precision of this measurement system can be very high (thousandths of a millimeter), but it depends on the angle of the vertex opposite the scanner (the further away from 90° the lower the precision), which limits the size of the scene to be analyzed. Since this angle depends strongly on the distance between the laser emitter and the camera, increasing the range means greatly increasing the size of the measuring equipment. In practice, the maximum range of these scanners is limited to 20-30 cm.
In most cases, instead of a measurement point, a line is projected that sweeps the surface of the object to speed up the acquisition process.
Some examples of 3D scanners by triangulation:.
The National Research Council of Canada was one of the first institutes to develop the technology on which triangulation scanning is based in 1978.[2].
This third type of scanner measures the phase difference between the emitted and received light, and uses this measurement to estimate the distance to the object. The laser beam emitted by this type of scanner is continuous and modulated power.
The range and precision of this type of scanner is intermediate, positioning itself as a solution between the long range of time-of-flight devices and the high precision of triangulation scanners. Its range is around 200 m in low noise conditions (low ambient lighting), and its characteristic error is around 2 mm per 25 m.
In some models the range is limited precisely by its mode of operation, since by modulating the beam with a constant frequency, there is ambiguity in the measurement of the distance proportional to the wavelength of the modulation used.
The precision of the measurement also depends on the frequency used, but in the opposite way to the range, which is why these concepts are complementary, and a compromise point must be found between the two, or two different frequencies must be used (multi-frequency-ranging or MF). In this way, using several modulation frequencies, the highest frequency will be used to calculate the distance to the point, and the lowest will be used to resolve the ambiguity of said measurement.
The acquisition speed is very high, with current models achieving scanning speeds ranging between 100,000 and 1 million points per second, depending on the required precision.
Summary of features:
Some examples of scanners based on phase difference:.
It is an interferometric technique by which a beam reflected from a surface passes through a birefringent crystal, that is, a crystal that has two refractive indices, one ordinary and fixed and another extraordinary one that is a function of the angle of incidence of the ray on the surface of the crystal.
As a result of passing through the glass, two parallel rays are obtained that interfere using a cylindrical lens. This interference is captured by the sensor of a conventional camera, obtaining a pattern of stripes.
The frequency of this interference determines the distance of the object at which the beam was projected. This technique allows the measurement of holes in their collinear configuration, achieving precision better than a micron. The advantage of this technique is that it allows the use of non-coherent light, this means that the lighting source does not have to be a laser, the only condition is that it be monochromatic.
The applications of this technique are very varied, from reverse engineering to the inspection of surface defects in the steel industry at high temperatures. Conoscopic holography sensors are manufactured by Optimet Archived February 2, 2011 at the Wayback Machine.
Conoscopic holography was discovered by Gabriel Sirat and Demetri Psaltis in 1985.
Structured light 3D scanners project a light pattern onto the object and analyze the deformation of the pattern produced by the geometry of the scene. The model can be one-dimensional or two-dimensional. An example of a one-dimensional model is a line. The line is projected onto the object being analyzed with an LCD projector or laser. A camera, slightly offset from the model projector, looks at the shape of the line and uses a technique similar to triangulation to calculate the distance of each point on the line. In the case of the single line model, the line is swept across the panorama field to gather distance information one strip at a time.
An example of a two-dimensional model is a grid or line model. A camera is used to record the deformation of the model and a fairly complex algorithm is used to calculate the distance at each point on the model. One reason for the complexity is ambiguity. Consider a series of parallel vertical laser streaks sweeping horizontally across a target. In the simplest case, one could analyze an image and assume that the left-right sequence of stripes reflects the succession of lasers in the series, so that the leftmost stripe of the image is the first laser, the next one is the second laser, etc. In non-trivial targets containing pattern change, pits, occlusions, and depth, however, this sequence is decomposed as streaks that are sometimes hidden or may appear even with changed order, resulting in laser streak ambiguity. This particular problem was recently solved by a breakout technology called Multistripe Laser Triangulation (MLT). Structured light scanning is still a very active area of research with much research published each year.
The advantage of structured light 3D scanners is speed. Instead of scanning one point at a time, they scan multiple points or the entire panorama field at once. This reduces or eliminates the problem of motion deformation. Some existing systems are capable of scanning moving objects in real time.[3].
See: Structured light scanner.
Modulated light 3D scanners emit a continuously changing light on the object. Generally the light source simply cycles its amplitude in a synodal pattern. A camera detects the reflected light and the amount the light pattern changes to determine the distance the light travels.
Passive scanners do not emit any radiation themselves, but instead rely on detecting radiation reflected from the environment. Most scanners of this type detect visible light because it is radiation already available in the environment. Other types of radiation, such as infrared, could be used as well. Passive methods can be very cheap, because in most cases they do not need particular hardware.
Stereoscopic systems use the same principle as photogrammetry, using the measurement of parallax between two images to determine the distance of each pixel in the image. They generally employ two video cameras, slightly separated, looking at the same scene. By analyzing 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 human stereoscopic vision.
These types of 3D scanners use sketches created from a succession of photographs around a three-dimensional object against a very well-contrasted background. These silhouettes are stretched and crossed to form the hull's visual approximation of the object. With these kinds of techniques some kinds of concavities of an object (such as the inside of a bowl) are not detected.
There are other methods that, based on the user's help in the discovery and identification of some characteristics and shapes in a set of different portraits of an object, are capable of building an approximation of the object itself. These kinds of techniques are useful for constructing the rapid approximation of buildings in the likeness of objects, formed and simple. Several commercial packages are available such as iModeller, Sculptor D or RealViz ImageModeler.
This type of 3D scanning is based on the principles of photogrammetry. It is also somewhat similar in methodology to panoramic photography, except that photos are taken of an object in three-dimensional space to replicate it rather than taking a series of photos of a point in three-dimensional space to replicate the surrounding environment.