3-D Imaging TerminologyOne of the documents to come out of committee E57 was E2544-08 -- "Standard terminology for three- dimensional (3-D) imaging systems." What follows is an excerpt from the document of some of the 3-D imaging terminology. To keep the excerpt short, we have included the definition of just a few of the terms listed.3.2 Definitions of terms specific to this standard 3-D imaging system--a noncontact measurement instrument used to produce a 3-D representation (e.g., a point cloud) of an object or a site.
Angular increment--the angle between samples, Da, where Da = ai- ai-1, in either the azimuth or elevation directions (or a combination of both) with respect to the instrument’s internal frame of reference
Beam diameter ( ds ) --for a laser beam with a circular irradiance pattern, the beam diameter is the extent of the irradiance distribution in a cross section of the laser beam (in a plane orthogonal to its propagation path) at a distance z and is given by: ds(z) = 4s(z) where: s(z) = sx (z) = sy(z) sx(z), sy(z) = the square roots of the second order moments
Beam divergence angles ( qsx , qsy ) --measure for the asymptotic increase of the beam widths, dsx(z) and dsy(z), with increasing distance, z, from the beam waist locations, z0x, and z0y, given by:
Beam propagation ratios ( Mx2, My2 )
Beam width ( dsx, dsy )
First return--for a given emitted pulse, it is the first reflected signal that is detected by a 3-D imaging system, time-of-flight (TOF) type, for a given sampling position, that is, azimuth and elevation angle
Flash LADAR
Instrument origin
Last return--for a given emitted pulse, it is the last reflected signal that is detected by a 3D imaging system, time-of-flight (TOF) type, for a given sampling position, that is, azimuth and elevation angle
Multiple returns
Point cloud--a collection of data points in 3-D space (frequently in the hundreds of thousands), for example as obtained using a 3-D imaging system
Registration
Second order moments ( sx2, sy2 )
Simple astigmatic beam
Spot size
Voxel--a discrete volumetric element in a 3-D grid representation of data
|
Applications for 3-D imaging systems span many disciplines and include generation of 3-D models, surveying and mapping, reverse-engineering, quality control, forensics, autonomous vehicle navigation, historical preservation, and archaeology. The use of 3-D imaging systems has increased rapidly in recent years; however, standard performance evaluation/calibration protocols have yet to be developed, and the need for standards to evaluate these systems has become critical.
The following statements, taken from a National Institute of Standards and Technology (NIST) U.S. Measurement System Task Group’s white paper, are applicable to all aspects of 3-D imaging systems:
“The current strategic environment, in which measurements are being conducted in both the public and private sectors, is characterized by rapid changes in technologies, the continued emergence of new measurement areas and nontraditional measurement needs, and increasingly blurred lines between disciplines accompanied by increasingly blurred lines between the national and international measurement systems.… The rate of change in new technologies is in some cases outpacing the ability of national and international bodies to come to consensus on needed support (standards, mutual-recognition arrangements, etc.) and put the appropriate infrastructures in place, leading to the emergence and proliferation of private groups establishing standards, measurements, and certifications, with or without governmental and international buy-in.”
Most manufacturers and users agree on the need for some form of standardization in terms of commonly used terminology, test protocols, targets/artifacts, and data-transfer protocols. For users, standards ensure that an instrument meets certain criteria. For manufacturers, standards allow for equivalent assessment of instruments.
With regard to terminology, users’ decisions to purchase an instrument are based, in part, on the manufacturer’s specifications. Confusion as to the definitions or usage of common terms such as “accuracy,” “resolution,” and “sampling speed” may lead to the purchase of an instrument (a large capital expense) that may not be the best suited for the intended application. From a manufacturer’s viewpoint, different definitions of the terms may lead a potential buyer to incorrectly conclude that one instrument is better than another.
Many of the terms such as “accuracy,” “uncertainty,” and “repeatability” already have universally accepted definitions that were developed by standards organization such as the International Organization for Standardization (ISO), American National Standards Institute (ANSI), and American Society for Testing and Materials International (ASTM). However, it should be noted that although these standard definitions have been in existence for many years, there still exists some confusion in the marketplace as to their meaning or their usage. The commonly used term “accuracy” is a good example.
“Accuracy” is defined by ISO as the closeness of the agreement between the result of a measurement and a true value of the measurand, and is a qualitative concept. However, the term “accuracy” has often been used interchangeably with “precision” and is assigned a value. Sometimes, the term “resolution” has also been used to mean “accuracy.”
Other terms commonly used in the 3-D imaging system community that have created confusion include “spot size” (Is it the diameter or the radius?), “resolution,” “scan speed,” “range,” and “control points.”
Besides standard definitions, standard test protocols and reporting of results are also required to allow for fair comparisons of instrument capability. Let’s assume, for example, that the agreed-upon definition for error is the difference or deviation of the measurement from the “true” or accepted reference value. The adoption of a definition alone, however, is not sufficient to give a buyer a sufficient basis for making a rational comparison. If the stated manufacturer accuracy is ±1 meter, can a buyer assume that if a target is located at 100 meters, the measurements will be between 99 and 101 meters 100 percent of the time? If so, how would one verify this? The obvious answer would be to conduct a performance evaluation or “calibration” of the instrument that requires test protocols. In this case, standard protocols are desirable because different test protocols will likely yield different results. The test protocols should detail the methodology and how the results are reported.
In addition to knowing the specifications of an instrument, users also need to know if an instrument is suitable for their particular application. For example, if the 3-D imaging system is going to be used for recording historic structures or works of art, the ability of the scanner to capture small features is important. If the instrument resolution is defined as the smallest distance--in the range (depth), horizontal, and vertical directions--between two objects that is discernible by the instrument, then standard procedures have to be developed on how to determine the instrument’s resolution in all three directions. A possible method would involve the use of artifacts; thus, standard artifacts (size, shape, material, etc.) will have to be developed as part of the methodology. Again, standardization of these procedures would allow for accurate comparisons between instruments.
In situations where there is potential for litigation, the existence of standards is crucial in substantiating one’s findings or results. These situations include the use of 3-D imaging systems for forensics, surveying, as-built modeling, and determining billable costs for excavation and dredge volumes.
To address the lack of standards for 3-D imaging systems, an ASTM committee for 3-D imaging systems, ASTM E57, was established in 2006. The committee was formed specifically to develop standard terminology, test methods, best practices, and data interoperability specifications for these instruments. The scope statement for E57 is as follows:
“The development of standards for 3-D imaging systems, which include, but are not limited to, laser scanners (also known as LADAR or laser radars) and optical range cameras (also known as flash LIDAR or 3-D range cameras).
The initial focus will be on specification and performance evaluation standards for 3-D imaging systems for applications including, but not limited to, construction and maintenance, surveying, mapping and terrain characterization, manufacturing (e.g., aerospace, shipbuilding, etc.), transportation, mining, mobility, historic preservation, and forensics.”
Standard test methods for the performance evaluation of 3-D imaging systems will provide a basis for fair comparisons of instruments, provide assurance that an instrument will meet certain criteria, reduce the confusion regarding terminology, and increase user confidence in the applications of these systems. Best practices will improve product quality and will ensure a minimum level of performance. Data interoperability will ensure the seamless transfer of 3-D image data through the use of open standards and data-exchange formats. A cost analysis associated with the lack of interoperability may be found in the report “Cost Analysis of Inadequate Interoperability in the U.S. Capital Facilities Industry,” by Michael P. Gallaher, Alan C. O’Connor, John L. Dettbarn, Jr., and Linda T. Gilday (National Institute of Standards and Technology, 2004).
Current subcommittees of ASTM E57 are working on standards for terminology (E57.01), test methods (E57.02), best practices (E57.03), and data interoperability (E57.04). Full committee meetings are held annually in January and in June. Subcommittees meet more frequently.
ASTM E57 published its first standard in 2007, titled ASTM E2544-08--”Standard terminology for three-dimensional (3-D) imaging systems.” (See sidebar on page 36.) The terminology subcommittee continues to add new terms to this standard. The test methods subcommittee is working on a method to evaluate the ranging performance of 3-D imaging systems. The best practices subcommittee is working on a document on the safety requirements for 3-D imaging systems. The data interoperability subcommittee is developing a 3-D image data format requirements document.
Because the standards and documents produced by ASTM E57 can have wide-ranging effects--for example, how manufacturer specifications are written, how customers decide which instruments to purchase, and how contractual language for 3-D imaging services are specified--input and participation from the larger 3-D imaging community is actively sought.
Geraldine S. Cheok joined the NIST Structures Division in 1984.
Since 1998, she has been with the Construction Metrology and Automation
Group (CMAG). Her work has involved the use of 3-D imaging systems
(e.g., laser scanners, LADAR, laser radars, range cameras) for
construction applications.
Alan M. Lytle joined CMAG in 2001 as a robotics engineer. His
current work includes research toward automated construction, on-site
component tracking, and construction object recognition using 3-D
imaging systems. He is the current chair of the ASTM E57 committee.
Kamel S. Saidi, Ph.D., is currently a guest researcher to NIST with
the building and fire-research lab. He is also involved in research
aimed at establishing standards for the performance evaluation of 3-D
imaging systems.
© 2023 Quality Digest. Copyright on content held by Quality Digest or by individual authors. Contact Quality Digest for reprint information.
“Quality Digest" is a trademark owned by Quality Circle Institute, Inc.
