In this paper we establish the need for communication between organizations involved in the fight against counterfeiting crime. We also examine the paradox in providing information that could serve the criminal as well as those attempting to protect themselves from criminal activity. Counterfeiting is estimated to account for over 5% of world trade. It is a global operation with no respect for international borders. It is increasingly sophisticated and increasingly the province of organized crime, which applies the techniques developed for drug distribution to the production and distribution of counterfeit articles. To fight this crime there is an increasing plethora of authenticating features and technologies available. Many companies do not recognize the problem and the number of anticounterfeit technologies can be confusing for potential users. There is therefore a need for information about them, their comparative characteristics, to be easily available. At present there is inadequate communication between those who develop and produce anti-counterfeiting devices and those who use them, notwithstanding the marketing efforts of the former. Communication which stimulates and encourages the spread of information between those engaged in the fight against counterfeit crime can only help in that fight. But what we term 'the communication paradox' requires circumspection and care in the content and the distribution of such information. The communication paradox is that the better the channels of communication, the easier it is for criminals to get hold of that information. The challenge is to institute communications which are effective but restrictive. More communication of information between those engaged in counterfeit deterrence will enhance individual companies' and organizations' anticounterfeit efforts and thus contribute to an overall improvement in the fight against counterfeit crime.

A method of security design is presented, founded on a systematic seven step approach: (1) drafting a security policy, (2) a list of requirements, (3) a security scheme, and (4) a fraud-risk analysis. Based on these documents: (5) the security measures can be selected, (6) and implemented, (7) to result in a security product. Simulation and/or experiments are required to assess the relevant properties of the design or product. The actual evaluation consists of comparing these properties with the requirements. Subsequently weak and strong points, and possible paths of attack will be established. In case the evaluation reveals a considerable mismatch between requirements and characteristics, the design cycle has to be passed through again and either the requirements must be adapted, or the design, or both. This procedure is repeated until the remaining mismatch between requirements and design properties becomes acceptable. It is paramount that an evaluation is not postponed until the final stage of product realization, but is carried out in the early phases of the project. In fact it is beneficial if all seven step of the design are the subject of an evaluation. Some design considerations for optically variable devices are discussed.

Security evaluations are carried out for several reasons. They can be used for product improvement, in a risk analysis, or they can be used to verify the security claims or targets. Using a simplified design process scheme it is explained where, and to what extent security evaluations can be useful. Depending on the anticipated risk, or the trust the parties involved (manufacturers, users, or issuers) have in certain security related products, different sorts of evaluation can be carried out. The goals and the assumptions made for the different sorts of evaluation are discussed, as well as the relation between the sorts of evaluation. An important tool in evaluating security systems is the matrix security model. This model is discussed in some detail.

Features for use in the first new design of United States currency in more than 60 years were selected from a field of over 130 possibilities. To narrow the field, individual features were subjected to a variety of evaluation methods, ranging from subjective to rigorous physical to limited production testing. Comprehensive evaluations were carried out by an independent committee of experts through a Treasury contract with the National Research Council (NRC) as well as by the interagency New Currency Design (NCD) Task Force of technical experts from currency production, issuing, processing and enforcement authorities. Features were also rated for effectiveness, durability, feasibility, intelligibility, security, production and processing impact and compatibility with design requirements. Through successive levels of evaluation, the original field was narrowed to 35 for scale-up testing to a handful for production testing. Full production testing and implementation schedule determined the final mix of features included in the new design Series 1996 $100 note.

The first new design of United States currency in over 60 years will soon be issued. Its issuance will be the culmination of a 6-year effort to make U.S. currency more secure against widely available advanced reprographic technology. The cooperative effort was directed by the Advanced Counterfeit Deterrence (ACD) Steering Committee, with executive representatives from the Federal Reserve System (FRS), U.S. Secret Service (USSS), Bureau of Engraving and Printing (BEP) and Treasury Department. A task force of technical experts from each agency carried out the necessary evaluations. The overall strategy to determine the new design and new features applied a comprehensive, synergistic approach to target each type of currency user and each type of counterfeiting. To maximize objectivity yet expedite final selection, deterrent and detection technologies were evaluated through several parallel channels. These efforts included an open request for feature samples through the Commerce Business Daily, in-house testing of each feature, independent evaluation by the National Research Council, in-house design development and survey of world currencies. Recommendations were submitted by the Steering Committee to the Treasury Secretary for concept approval, announced in July 1994. Beginning in 1996, new designs will be issued by denomination approximately one per year, starting with the $100 bill. Future new design efforts will include input from the recently founded Securities Technology Institute (STI) at Johns Hopkins Applied Physics Laboratory. Input will include evaluation of existing features, development of new techniques and adversarial analysis.

The estimated 3.7 million Americans with low vision experience a uniquely difficult task in identifying the denominations of U.S. banknotes because the notes are remarkably uniform in size, color, and general design. The National Research Council's Committee on Currency Features Usable by the Visually Impaired assessed features that could be used by people who are visually disabled to distinguish currency from other documents and to denominate and authenticate banknotes using available technology. Variation of length and height, introduction of large numerals on a uniform, high-contrast background, use of different colors for each of the six denominations printed, and the introduction of overt denomination codes that could lead to development of effective, low-cost devices for examining banknotes were all deemed features available now. Issues affecting performance, including the science of visual and tactile perception, were addressed for these features, as well as for those features requiring additional research and development. In this group the committee included durable tactile features such as those printed with transparent ink, and the production of currency with holes to indicate denomination. Among long-range approaches considered were the development of technologically advanced devices and smart money.

The criteria by which optically variable devices are judged are aesthetic, semantic, security, ergonomic, and physical/chemical. This paper addresses ergonomic aspects which relate to the human vision and perceptual-cognitive system. Applying some pertinent rules may help greatly to improve the image visual information for easier, more straight-forward reception of a persistent security message. We consider two important aspects of the human visual system that help to determine the ergonomic response to visual displays created using optical diffraction. The human visual system aspect treats the retinal source of information, which is the retinal signal produced when an image of the external world is projected on the retina. The other aspect is the underlying information-processing mechanism of our brains and its constructive operations, which yields the final perceptual information. In this paper we consider information processing methods hidden in the biology of our cognition system. Findings on the relationship between physiology and psychology, sensory results and the activities of the optic pathway and subjective brightness sensations can be applied directly in designing images. Some effects are demonstrated by video tape.

This paper discusses the incorporation of an opto-biometric method into identification documents. Factors important to a successful implementation strategy are discussed. Design follows from anticipated use of the document, as well as size, cost and time factors. Public sector documents face additional privacy concerns which contribute psychological and emotional factors to the choice of opto-biometric features. Travel documents based on international standards face further constraints based on national laws and customs.

The new Dutch passport incorporates a series of security features. Why these features were chosen, what they look like and what purpose they serve are discussed in this paper.

Since their introduction at the SPIE-conference 'Holographic Optical Security Systems' in The Hague in 1991, the printed security structures like SAM (screen angle modulation) and FREM (frequency modulation) have proven their value in the protection against digital color-copiers. Over the past years the SAM security structures have been used extensively in security documents, not only as a anti-copying structure but increasingly as an optically or machine detectable security feature, using respectively a SAM-screener and Laplace-filtering. This last feature of optical- or machine-detectability is becoming of more importance due to the fact that publicly available image processing software is getting more and more advanced. Once a document has been scanned into a computer, the use of digital filtering techniques may well overrule, imitate or simulate many security features. By increasing the frequency of the SAM- structures to a higher level, leading to so-called (mu) SAM (micro SAM) or the use of sample band image coding (SABIC), structures are obtained that cannot be reproduced with standard equipment. Moreover, these structures have been designed such that their presence in a security document is easily checked, either by simple optical means or during scanning by a simple image processor. For different application areas different techniques have been developed such as symmetry-breaking ((mu) SAM screener), (one-sided) Laplace-filtering, envelope-detection and optical Fourier domain-filtering.

Smart (intelligent) printing is the creation of useful patterns beyond alphanumerics and graphics immediately obvious to the human eye. It employs smart inks, patterns, surfaces and substrates. Recent proliferation of color copiers, personal computers and scanners has facilitated a tenfold increase in counterfeiting in many countries over the past three years. Banknotes, cheques, academic certificates, art work, visitors passes, venue tickets and many other artifacts have been compromised. Paradoxically, the best counterfeits produced by some foreign governments and organized crime are rarely the main problem. The secret services of many countries use forensic science to great effect in pursuing these fairly readily identified sources of limited number. Bad counterfeits usually made on color copiers or computers, with or without color scanners, are the most difficult to combat because they are made by very large numbers of casual counterfeiters who may never commit crime again. For instance, counterfeit banknotes intercepted by the Bundesbank have been photocopies in a fluctuating range of 50 - 84% of cases in the last four reported years. Cheque and other document fraud is also inflated by these burgeoning bad copies and here we must add amateurish alterations using copiers or scanners. For instance, a better academic degree can mean a better job, an interbank transfer form can be 'raised' in value by enormous amounts. The issuer of a 'bad' counterfeit does not mind that it is usually picked up on a second transferral. They are long gone by then or, with banknotes, they can deny that they issued it. First priority in reversing the upward trend of counterfeiting must not therefore be the creation of better secret features traceable by forensic laboratories over extended periods of time. Rather we need better and more obvious optically unique features, not easily emulated, that can be spotted in the split second when several, say, banknotes are handed over in a dimly lit surrounding. It is usually impractical for the recipient to use a portable optical or electronic checker. Nevertheless, better, cheaper, smaller and faster validating instruments would also be a help, particularly for small shops. Here the new Mars Electronics Cashguard banknote validator is great progress. It performs rapid complex analysis on banknotes yet costs well under $500. Designs must improve though. Advanced aliasing takes advantage of the fact that copiers and computer scanners have poor resolution and scan in a certain way. So far it has been useful on color documents: gray versions are particularly effective making words like ILLEGAL COPY appear on all copies. However, smart patterns such as Kalamazoo Copyvoid have been of less use against monochrome counterfeits -- photocopied expensive books, vehicle insurance forms and sheet music for instance. This is because the contrast controls can be used to wipe it out (with color the colors would be ruined by such action). However, the Kalamazoo Laboratories in the UK have just announced a new version of Copyvoid that works at both high and low contrast photocopying or scanning. Indeed, it is also milder and more even to look at so even sheet music can be printed over it using conventional inks and still read clearly. The problem of 'bad ' counterfeits is very severe with 1 in 100 counterfeit banknotes being suffered in some UK locations and Northern Ireland seeing a tenfold increase in counterfeits overall in the last 12 months. Cheque fraud doubles each year in some countries. The solution here must be for the authorities to totally redesign both far more often -- say at least every five years -- and follow best practice in totally withdrawing/invalidating the old ones.

This paper presents a process able to mark digital pictures with an invisible and undetectable secret information, called the watermark. This process can be the basis of a complete copyright protection system. The process first step consists in producing a secret image. The first part of the secret resides in basic information that forms a binary image. That picture is then frequency modulated. The second part of the secret is precisely the frequencies of the carriers. Both secrets depend on the identity of the copyright owner and on the original picture contents. The obtained picture is called the stamp. The second step consists in modulating the amplitude of the stamp according to a masking criterion stemming from a model of human perception. That too theoretical criterion is corrected by means of morphological tools helping to locate in the picture the places where the criterion is supposed not to match. This is followed by the adaptation of the level of the stamp at that place. The so formed watermark is then added to the original to ensure its protection. That watermarking method allows the detection of watermarked pictures in a stream of digital images, only with the knowledge of the picture owner's secrets.

The successful use of many security papers, foils and films depends on the technology of chemical fastening systems -- especially pressure sensitive adhesives. These are adhesives activated not by heat or by the evaporation of water or some other solvent, but simply by the act of application -- by pressure. These adhesives provide the means whereby laminations, substrates and seals are made effective. In addition to their physical properties these adhesives are often required to possess optical properties to allow the security materials to be visibly active and indeed the adhesive system may itself contribute as a carrier for a variety of security materials. Recent advances in adhesives chemistry have made it possible to achieve virtually all the required physical performance characteristics combined with a choice of optical properties ranging from total opacity to invisibility and including controlled translucency and tinting. The implications for security printing and packaging are important. Opacity is easy to achieve, for example by loading the adhesive with aluminum powder, by the selection of totally opaque materials like metallized film or by various printing processes. But achieving transparency is a different matter, and transparency is mandatory for applications involving the protection of documents, photographs, etc. with a clear film over-laminate. Obvious examples would be for passports, visas and other personal identification. But some security devices may themselves require protection; for example holograms or embossings. And transparency in the test laboratory is not enough. The Australian driving licence is stuck to the windshield, so the transparency of the adhesive must be sustained over long periods without deterioration due to prolonged u/v exposure, climatic conditions or aging. The commercial label market has helped to push the technology forward. There is a strong demand for the 'no-label look' for packaging of clear plastic and glass containers where the content can be easily seen without interference and where wording or symbols can be read through the container. You see this increasingly with pharmaceuticals, cosmetics and toiletries, even bottled beer. Achieving transparency is one thing but this property must be combined with all the physical properties required of the pressure sensitive adhesive. First there is the question of permanence, re-positionability and removability and the degree to which these features are required. Secondly many complications arise from the range of materials to which the adhesive must be anchored and the range to which it will be applied and must bond. Obviously these surfaces vary from those with the highest surface energy (polycarbonate for example) to those apolar surfaces engineered for minimum attraction (PTFE -- 'Teflon' for example).

In fraud prevention one of the most critical issues is how to distinguish a genuine document or object from a counterfeit. Most current technologies are based upon addition of a same, hard to counterfeit, addition to the document, like a hologram, a watermark or a graphical technique. Another approach is to add a unique property, which is different to each document. 3DAS is such a technique. 3DAS markers are all unique and can prove the authenticity of a document or object. They can also be used to identify objects and documents. This article explains the principles of 3DAS and informs the reader of the current status.

The advent of 'intelligent,' electronic data bearing tags is set to revolutionize the way industrial and retail products are identified and tracked throughout their life cycles. The dominant system for unique identification today is the bar code, which is based on printed symbology and regulated by the International Article Numbering Association. Bar codes provide users with significant operational advantages and generate considerable added value to packaging companies, product manufacturers, distributors and retailers, across supply chains in many different sectors, from retailing, to baggage handling and industrial components, e.g., for vehicles or aircraft. Electronic tags offer the potential to: (1) record and store more complex data about the product or any modifications which occur during its life cycle; (2) access (and up-date) stored data in real time in a way which does not involve contact with the product or article; (3) overcome the limitations imposed by systems which rely on line-of-sight access to stored data. Companies are now beginning to consider how electronic data tags can be used, not only to improve the efficiency of their supply chain processes, but also to revolutionize the way they do business. This paper reviews the applications and business opportunities for electronic tags and outlines CEST's strategy for achieving an 'open' standard which will ensure that tags from different vendors can co-exist on an international basis.

Experiments on an optical pattern recognition system used for validation and security verification are provided. The system verifies a random phase mask bonded to a gray scale primary image. Using a limited set of images, the system performance is investigated in the presence of different types of input noise and distortions such as scratches, fingerprints, and bendings. The experimental results indicate that for the noise and the distortion tested here, the system provides a good correlation performance by verifying the phase mask.

An edge-lit holographic optical element has been developed which produces a high contrast and high resolution fingerprint image. Continuing work on the development of a holographic optical element for use in a compact fingerprint imaging system is described. The hologram is combined with an electronic imaging array into a package only several millimeters thick, using no image transfer lens system. Previously reported results with green holograms are reviewed. New results are presented for red holograms illuminated by a laser diode. Design issues and experimental results are discussed, relating to selection of hologram operating wavelength, angular and wavelength sensitivity of the hologram, and image quality versus packaging volume tradeoffs.

Optical variable devices are now widely used on documents or values. The most recent optical visual features with high definition, animation, brightness, special color tune, provide excellent first and second levels of authentication. Human eye is the only instrument required to check the authenticity. This is a major advantage of OVDs in many circumstances, such as currency exchange, ID street control . . . But, under other circumstances, such as automatic payments with banknotes, volume ID controls at boarders, ID controls in shops . . . an automatic authentication will be necessary or more reliable. When both a visual and automated authentication are required, the combination, on the same security component, of a variable image and a machine readable optical element is a very secure and cost effective solution for the protection of documents. Several techniques are now available an can be selected depending upon the respective roles of the machine readability and visual control.

This paper reviews four optical inspection systems, in which development TNO Institute of Applied Physics was involved: (1) intaglio scanning and recognition, (2) banknote quality inspection, (3) visualization and reading of a finger pattern, and (4) 3DAS authentication. (1) Intaglio is reserved for high security printing. It renders a tactile relief that can be recognized by a laser scanning technique. This technique is applied by various national banks to detect counterfeit banknotes returning from circulation. A new system is proposed that will detect intaglio on arbitrary wrinkled banknotes. (2) A banknote fitness inspection system (BFIS) that inspects banknotes in specularly reflected light is described. As modern banknotes are provided increasingly with reflective security foils, a new system is proposed that inspects banknotes in specular and diffuse reflection, as well as in transmission. (3) An alternative visualization method for visualization of finger patterns is described, employing a reflective elastomer. A CD scanning system reads the finger patterns. (4) A nonwoven structure has two advantageous properties for card authentication: a random structure which renders each few square millimeters of the pattern uniqueness (identification) and a 3D structure which makes it virtually impossible to be counterfeited (authentication). Both properties are inspected by an extremely simple lenseless reader.

In this report we discuss the method of the analysis of pseudo-identical optical signals. Two similar images, composed by identical apertures, were taken as an example for examination. The difference in their structures can be connected with the value of the apertures' function of transmittance, with the inter-apertures' separation along one direction, or with both these factors simultaneously. The overall functions of the transmittance for this type of pseudo- identical structure can be measured up through comparison of the secondary Fourier spectrums of corresponding transmittance functions.

OVDs are increasingly used as a means of authenticating documents or reducing the chances of successful counterfeiting. Kinematic effects are those observed when certain image elements appear to move on rotation or tilting of the substrate. Examples of devices showing these effects are described as 'Kinegrams' or 'Exelgrams' or, more recently, 'Kineforms.' All are produced by the creation of microscopic diffracting structures. All have in common: graphical animation, well defined mutual movements, an abandonment of holographic three dimensionality, and the ability to perform under diffuse lighting conditions and under most viewing angles. The images are not based on objects but are computer generated. Their principal advantage is that they can only be produced using highly sophisticated manufacturing techniques. These techniques are radically different from each other, and it is therefore surprising that the observed effects are so similar. This paper attempts to describe the observed features of kinematic OVDs in a structured way and to assess their relative performance on a micro as well as a macro scale, using a mixture of optical and scanning electron microscopy. It is hoped that this work will contribute towards a system of classification of KOVDs which will assist prospective users in the specification of such products.

We consider several aspects of security properties of holographic optically variable devices that are used for document protection. Features of optically recorded rainbow holograms, stereograms, holograms with pixel structure, and volume holograms are discussed from the point of view of optics. Similar approach is used with respect to optical properties of computer generated stereograms and digital holograms with pixel structure. Optical properties which are considered include diffraction efficiency, uniformness of the recording, angular range of horizontal parallax, multiplicity of recording, spatial separation of recording planes, color visual perception, spectral range, as well as transmission and reflection properties. The second group of security properties is connected with employed technology. Technological aspects of a hot stamping foil structure, self-destructive labels, and other are considered. Security properties of holographic optically variable marks are also analyzed in terms of novelty of employed techniques, the level of scientific and technological complexity, as well as resistance to imitating.

The most frequently used optical security features to protect documents and goods against counterfeiting are various types of rainbow holograms and diffractive optical variable devices (OVDs). Although these features offer a fair degree of security and most known falsifications are rather crude imitations it is still possible to copy them with all hidden and visible information. In our opinion it would be helpful to have copy proof holograms containing a machine readable feature to achieve a doubtfree verification at point of sale and other sites. We present our work towards the development of a copy proof and machine readable hologram with an optically encoded security feature.

In security printing in general and currency printing in particular, it is very rare for a single security feature to carry the total burden of 'authenticity.' There is almost always a layering of features, which in their complexity and coherence eventually determine the genuine. We give examples to show holography is no exception to the rule of layering. After discussing the integration of various layered holographic features into a security product, the paper discusses currency, proposing authentication of currency is, today, fundamentally a human perception and that the human eye is more attuned to recognition of a face than a machine readable random pattern.

We demonstrate the use of diffractive optical memories for official documents, such as machine-readable identity or fiduciary papers. Through engineering of the diffractive micro- structures, the direction and intensity distribution of the diffracted light can be tailored to optical memories for high security, uniqueness and unambiguous verification. The proposed optical memory is of the WORM-type, that is, write-once, read-many times. In order to write in the optical memory, the diffractive structure is changed irreversibly through the interaction of the diffractive surface with a beam of laser light. We demonstrate optical memories based on diffractive structures with a memory capacity of up to 100 kBits/cm² which are appropriate for use in securing official documents.

Optically variable devices (OVDs) are relatively new security features which are currently finding widespread application on a variety of security documents as a means of protection against counterfeiting. The OVD is in general a complex optical recording and the commonest form seen today is based on the presence of optically diffracting features, which are manufactured using embossing technology. This presentation deals with one particular type of security product -- a transparent or semi-transparent document overlay which may include an OVD combined both with UV-fluorescent or other special links, and may also include individualized information applied by laser-engraving technology. The main applications lie in the field of paper-based security documents such as passports, visas, driver's licenses and ID- cards.

A new optically variable device characterized by both rainbow holographic effects and two kinds of hidden elements is presented. Both components are recorded as computer generated Fourier holograms. The first hidden element is a slit or small image formed by a line of text or other image contained in a slit-shaped frame. Under normal illumination conditions the first element shows rainbow holographic and kinematic effects. The second element is a darker image visible at an angular range different than that of the first element. This second image is less visible under usual white light illumination conditions, but appears sharply in laser light. The advantage of the optically variable device proposed is that it gives both rainbow holographic effect and hidden Fourier-plane images at the same time from the whole hologram area. In this way, the need to use special encoded subfields which attract attention is eliminated.

The current use of optically variable devices (OVDs) is based on the well-known technology of hot-foil stamping that allows security features to be transferred to documents. Different types of OVDs such as holograms, kinegrams, pixeigrams or blazed gratings can be applied by this way. In spite of progress in creation computer-driven hot-foil transfer machines based on dot-matrix thermoprinting, the most typical technique utilizes presses with special stamps. A novel approach to hot-foil transfer involves the application of foil without temperature sensitive adhesive. Such an adhesive is distributed imagewise directly on the embossed hot-foil surface. A simple flat stamp or a laminator can be used for the application of different foil images. The said method gives more flexibility to customers in creating their own security features. The features may contain variable information: numbers, text, pictures and drawings of any kind. One of the main problem of traditional hot-foil transfer is the restricted resolution that strongly depends from stamp quality. The described process simply needs a flat stamp and its resolution mainly depends from initial adhesive distribution. The currently achieved resolving power is about 250 dpi and this figure can be further improved.

Optically recorded Fourier holograms readable by special laser detectors are frequently used as 'hidden' elements on optically variable devices (OVDs) used for security purpose. However, they are not difficult to identify and reproduce in holographic laboratory. In this paper we propose the use of the phase only computer generated Fourier hologram for security purposes. Its main feature is a possibility of recording a referenceless Fourier hologram which may form a nonsymmetric image. Such an element may be viewed even in the sunlight and it is impossible to manufacture it using standard analog holographic techniques. To obtain the best quality the hologram should be calculated using iterative Fourier-transform algorithm (IFTA) and manufactured using multiple-phase-step or continuous-phase technologies.

In some cases, it is desired to produce images that are suitable for one application and unsuitable for others. With the 'Vatican Library Accessible Worldwide' and 'Luther Digital' projects, for example, it was desired to make available through the Internet images of Vatican Library and Luther Library manuscripts that are suitable for scholarly study yet are unsuitable as a source for unapproved publication. One of the techniques used to accomplish this is the visible image watermark. Our technical goals for watermarking include (1) applying a readily visible mark to the image that clearly identifies its ownership, (2) permitting all image detail to be visible through the watermark, and (3) making the watermark difficult to remove. In this paper, we describe the technique we use. We also discuss the characteristics of good watermarks and options we have used in their application. illustrative watermarked images are presented.