Terms such as resolution, sensitivity and signal-to-noise ratio are often discussed by vendors and users without clear understanding by either party of just what they mean. These terms were already difficult enough to understand and apply when only RS-170 scanning was in broad use and when video was obviously an analog signal produced by a vidicon image tube.
Now, with a multiplicity of scanning modes in use, digital video transmission of many different formats and rates and an enormous variety of image sensing devices, both tube and monolithic, available, the need for a common vocabulary, universal test methods and a set of broadly-applicable interface specifications in the optical, mechanical and electronic domains is badly needed.
In this review, we will summarize current efforts to produce appropriate standards documents, assess the progress of these efforts, recommend priorities which might be followed to maximize early utility and identify additional standards efforts which should be undertaken.
This critical review covers ANSI/AIIM MS44-1988, American National Standard Recommended Practice for Quality Control of Image Scanners, and related standards both approved and in development.
The introduction of optical storage systems produced a problem: original documents were being routinely destroyed after scanning, and there was no way, short of examining each electronic image, to determine whether the electronic images were legible. This problem was recognized by vendors and by users in both the commercial and government arenas. These groups requested that the Association for Information and Image Management (AIIM) produce a standard. This review covers the content of that standard, MS44, and the process of writing it.
The principal aim of MS44 is to explain to the unsophisticated end user the need to implement quality control procedures, and to provide simple, usable procedures and tools. This was done by providing test targets with instructions for their use. The targets allow the user to test the scanner’s image resolution, linearity, reproduction of various type sizes and halftone screens, and color blindness.
MS44 was produced within two years, which is very fast for a national standard. This review suggests ways in which the progress of other standards might be accelerated.
When I was asked to prepare this paper, my immediate response was that the topic was overwhelming. However, the more I thought about the issues the more I realized that because we understand so little about color and electronic imaging, I could get away with almost anything. I won’t take that much freedom.
Although image processing has a longer history than computer graphics, an imaging standard has never been established. An ad hoc imaging task group had the first meeting in January 1988, and an ANSI X3H3.8 imaging group was formed thereafter. The task group is chartered to standardize the existing imaging technology. This effort has attracted many companies to join the task group, and great progress has been made since. Meanwhile, the International Standard Organization (ISO) also started its effort last year for establishing an international imaging standard. Due to this new development at the ISO level, ANSI has decided to merge its effort with ISO, and establish one imaging standard.
An international team of technical experts, the Joint Photographic Experts Group (JPEG), has been working for the past four years on a color image data compression standard for use worldwide in a variety of still image applications. This joint committee operated under ISO/IEC JTCl/SC2/Working Group 8 (Coded Representation of Picture and Audio Information) until it recently moved to a new working group, ISO/IEC JTC1/SC2/WG10 (Photographic Image Coding). The work is being done jointly with a special rapporteur group on image communication within CCITT Study Group VIII.
Within the United States, work in support of this international activity is focused within the U.S. picture coding committee, Accredited Standards Committee X3 Technical Committee X3L3 (formerly Task Group X3L2.8). Other standards applications groups are eagerly awaiting the completion of JPEG in order to effectively incorporate color images into their application environments.
JPEG's intensive technical research and refinement stages have been completed and an International Standards Organization (ISO) Committee Draft for the evolving JPEG data compression standard has been submitted for international balloting. Extensive testing of the applicability to various environments is in progress. Companies are beginning to announce JPEG software or hardware availability, even though they recognize that the JPEG algorithm may change.
This paper will provide a critical review of the evolving JPEG color image data compression standard. A history of JPEG is given first, followed by the steps still needed for adoption of JPEG as an international image data compression standard. Then the JPEG "toolkit" is explored with additional detail on the entropy coders and data interleave conventions. In the results section, the compression performance numbers on the JPEG test images are given for the baseline Huffman sequential coding system both with and without resynchronization. The arithmetic coding results are given for the sequential DCT-based mode with and without resynchronization, several progressive DCT-based processes and the sequential lossless mode. The final section briefly notes some applications and implementations of JPEG.
Although facsimile systems have existed for more than twenty five years, only recently have we seen explosive growth in the use of facsimile, with a corresponding increase in facsimile standards activity.
Current standards work concentrates on improving the quality and speed of image transmission, and on increasing functionality. The goal is to benefit from the developing technology - higher resolution scanners and printers, and improved processing, storage and communication technology. Topics now under discussion include (but by no means are limited to) transmission of color images, improved compression of bi-level images, facsimile routing, facsimile carried by the Integrated Services Digital Network (ISDN), and facsimile applied to teleconferencing.
Personal computers can operate as or can be used to add functionality to facsimile terminals. The result is new standards activity (for example, binary file transfer, data base storage and retrieval, computer-facsimile communication, and an Applications Programming Interface).
This paper describes the facsimile and facsimile- related standards activity with particular emphasis on function and performance. Image compression algorithms are an important factor. Communication protocols are also addressed.
There are several important international standards in development that have implications for Flat Panel Display Technology. There are ergonomic standards directed toward the display work place and the flat panel standards directed toward the technical interface between suppliers of flat panel subassemblies and manufacturers of systems that employ flat panels as a principal or ancillary electro-optical device.
The trend in the ergonomic case is to standardize the work place. Therefore, the standards are developed with a view to homologate the union of the environment, the equipment, the application and the end user. Since, to date, more than 10 CRT displays are in the office work place for every flat panel, it is natural that the body of ergonomic data upon which these standards are based anticipate the use of the CRT.
Since no up-to-date standards for measurement and description of flat panel displays exist, a work station developer cannot rely on specification sheets from different flat panel manufacturers for decision making data. The industry has still not come to view itself as required to provide a valuable convenience over CRT (portability or unique space utilization) while imposing as few compromises in function as possible.
This paper will discuss those elements of emerging international standards which are key, and outlining the implications for various technologies. The implications will be projected into the future to address: “What has to change for flat panels to compete in the 90s?”.
It has been the consistent pattern in the development of any new imaging system that permanence is not one of the initial properties to be considered by the manufacturer. This is very understandable behavior. Prime consideration during product development must be directed towards the successful operation of the new system. Ironing out flaws, improving reliability, determining means of production and obtaining sources of raw materials are all additional concerns that must be addressed. Another consideration is the reluctance of evaluation laboratories to be involved in long-time incubation testing on materials which are still undergoing change. The consequence, however, is that the ultimate life of the system is frequently not known even after it is on the market. This is reflected in the current status of standards and specifications on the preservation of magnetic media and optical disks.
Medical image quality is ultimately gauged by the image’s contribution to an accurate diagnosis. Meaningful comparison of the key elements of diagnostic image quality requires that consistent methods be used to measure them. A NEMA subcommittee is drafting standards to evaluate an x-ray image intensifier (XRII) system—the XRII; lenses linking the XRII output to a camera; video, photofluorographic, and cine cameras; and a video monitor. Image quality parameters include contrast ratio, detective quantum efficiency (DQE), fixed pattern noise, gain, image signal uniformity, limiting spatial resolution, modulation transfer function (MTF), noise power spectrum, radial distortion, signal-to-noise ratio, and visible entrance field size. An IEC Working Group is dealing with the XRII itself and first addressed MTF and DQE. This paper reviews the work of these two groups developing standards for measuring the image quality parameters of medical XRIIs and systems.
The federal government is about to invest billions of dollars to develop multimedia training materials for delivery on computer-based interactive training systems. Acquisition of a variety of computers and peripheral devices hosting various operating systems and suites of authoring system software will be necessary to facilitate the development of this courseware. There is no single source that will satisfy all needs. Although high-performance, low-cost interactive training hardware is available, the products have proprietary software interfaces.
Because the interfaces are proprietary, expensive reprogramming is usually required to adapt such software products to other platforms. This costly reprogramming could be eliminated by adopting standard software interfaces. DoD’s Portable Courseware Project (PORTCO) is typical of projects worldwide that require standard software interfaces. This paper articulates the strategy whereby PORTCO leverages the open systems movement and the new realities of information technology. These realities encompass changes in the pace at which new technology becomes available, changes in organizational goals and philosophy, new roles of vendors and users, changes in the procurement process, and acceleration toward open system environments. The PORTCO strategy is applicable to all projects and systems that require open systems to achieve mission objectives.
The federal goal is to facilitate the creation of an environment in which high quality portable courseware is available as commercial off-the-shelf products and is competitively supplied by a variety of vendors. In order to achieve this goal a system architecture incorporating standards to meet the users’ needs must be established. The Request for Architecture (RFA) developed cooperatively by DoD and the National Institute of Standards and Technology (NIST) will generate the PORTCO systems architecture. This architecture must freely integrate the courseware and authoring software from the lower levels of machine architecture and systems service implementation. In addition, the systems architecture will establish how the application-specific technologies relate to other technologies. Further, a computer-based interactive training applications profile must be developed. This profile, along with the systems architecture derived as a result of the RFA, provides the basis for identifying the needed standards. NIST will then accelerate the development of these standards using, but not restricted to, existing standards activities within established standards forums.
The federal multimedia courseware effort has adopted the Interactive Multimedia Association (INA) Recommended Practices for Interactive Video Portability as the baseline for the migration of computer-based interactive training systems to an open systems environment based upon international standards. The PORTCO strategy includes an evolutionary migration to a standards-based, Open System Environments (OSE). An important aspect of this migration strategy is to move to open systems via stepwise evolution rather than via quantum leaps.
Another area of concern is that of infrastructure issues, such as maintaining and supporting the technologies required for computer-based interactive training. The federal multimedia initiative will use the RFA-based architecture to differentiate between those technologies that can be maintained and supported by existing infrastructure mechanisms and those that require new mechanisms. Existing infrastructure mechanisms will be used and where infrastructure mechanisms do not exist, the approach will be to place high priority on establishing the appropriate mechanisms. Establishing an infrastructure mechanism is a nontrivial task requiring sustained investment of resources.
Photographic and electronic imaging technologies are being used in a wide variety of information systems, in virtually every industry. The standards and related activities that provide some commonalty within this wide range of processes are being developed by a myriad of organizations. What we need is an organization to take a leadership role in coordinating the distribution of information about these standards and related activities. One approach would be to provide abstracts of the technical content and the availability of published standards, and a status report on other activities (standards under development, technical reports, and other projects). The ANSI ITSB (Image Technology Standards Board) has requested that such a database be established. Most recently, the International Standards Organization/International Electrotechnical Commission (ISO/IEC) Joint Task Force on imagery called for the development of a worldwide imagery database.
The proposed database would be developed using commercial, off-the-shelf hardware and software. The plan is to have standards developers submit their data in digital form. Our presentation will address our accomplishments, problems, and expectations regarding the development of this database.