To form an understanding of the environment in which non- intrusive detection and inspection technologies are required to operate, the Narcotic Detection Technology assessment Team has undertaken a series of field studies. These field studies have focused on the phenomenology, fate and behavior of narcotic residue in real world environments. The overall goal of the tests is to give Law Enforcement officers the ability to accurately differentiate between individuals involved in the smuggling process and individuals innocently contaminated with narcotics. The latest filed study in this series was conducted in Miami, FL in February 1996. The field study comprised several individual tests. The first was a contamination and transfer study which focuses on human contamination resulting from contact with actual kilos of cocaine and the mechanism by which this contamination transfers to surrounding objects, if at all. The second was a secondary contamination study which focused on determining the conditions under which cocaine contamination transfers from objects touched by individuals who handled narcotics to innocent passerbys. The third was a persistence study which focused on the persistence of cocaine contamination on people under a variety of conditions. An overview of the tests and their preliminary results will be discussed.

Customs investigation and drug enforcement services are interest in a rapid and reliable identification of smugglers and dealers. In contrast workplace testing and traffic controls are aiming at the detection of intoxicated persons via the determination of illegal narcotics in body fluids like urine or blood. DRUGWIPE is a pen size, test strip based immunochemical detector for narcotic contaminations on surfaces. It is extremely simple to apply and takes about two minutes to read test results without depending upon any further technical means. This paper describes the applicability of DRUGWIPE to identify drug smugglers or dealers as well as consumers. With respect to the situation and the initial suspicion the test indicates handling as well as consumption. In cooperation with the Institute for Legal Medicine in Munich suspicious drivers were examined with DRUGWIPE for the abuse of illegal narcotics. Test results from this test series are presented and compared with the results from the blood or urine analysis. The question whether the detected traces of illegal narcotics on the body surface of suspicious drivers are combing transpiration or external contamination are discussed.

When a vehicle is used to transport narcotic contraband material trace levels of that material can be found on surfaces of the vehicle, people associated with the vehicle and surface they contact. The detection of these trace levels can help to target vehicles associated with the smuggling of the contraband. A study to determine the typical levels of narcotic material that can be detected from these surfaces has been performed by personnel from Graseby, using a variety of drug materials. The size and packaging of the drug materials has been prepared to try to reflect that typically found in smuggling operations. These tests show that for all hard drugs easily detectable traces of drug material can be found on the vehicle, the proxy and secondary surfaces handled by the proxy. For detection of cannabis, the condition of the original material had a great bearing ont he reliability of detection.

To form anunderstanding ofthe environment in which non-intrusive detection and inspection technologies are required to operate, the Narcotic Detection Technology Assessment Team has undertaken a series of field studies. These field studies have focused on the phenomenology, fate and behavior ofnarcotic residue in real world environments. The overall goal ofthe tests is to give Law Enforcement officers the ability to accurately differentiate between individuals involved in the smuggling process and individuals innocently contaminated with narcotics. Thelatest field study in this series was conducted in Miami, FL in February 1996. The field study comprised several individual tests. The first was a Contamination and Transfer Study which focused on human contamination resulting from contact with actual kilos of cocaine and the mechanism by which this contamination transfers to sulTounding objects, ifat all. The second was a Secondary Contamination Study which focused on determining the conditions under which cocaine contamination transfers from objects touched by individuals who handled narcotics to innocent passerbys. The third was a Persistence Study which focused on the persistence of cocaine contamination on people under a variety of conditions. An overview ofthe tests and their preliminary results will be discussed. Keywords: Cocaine, cocaine contamination, cocaine transfer, cocaine persistence, primary contamination, secondary contamination

The Narcotics Detection Technology Assessment (NDTA) program is a series of studies conducted to evaluate illicit substance detection devices. The ability to effectively detect cocaine and heroin particles is directly related to the efficiency of a detection device's sample collection design. The NDTA tests are therefore structured to require sampling of narcotics from a surface. Tests standards are required which permit subnanogram to microgram quantities of narcotic to be dispensed onto a target surface for sampling. Optimally, the standard should not adversely affect the performance of the device under test. The NDTA test team has developed and experimentally characterized solution- deposited substrate standards, solution-deposited substrate- free standards, vapor-deposited standards, suspension standards, and dry mix standards, and dry mix standards. A variety of substrates and dry-mix fillers have been evaluated, including sand, fullerenes, copper powder, nickel powder, pulverized paper, and aluminum. Suspension standards were explored with a variety of liquids. The narcotic standards with the best performance were found to be dry mixes of cocaine with silver-coated nickel powder, and dry mixes of heroin with silanized glass beads.

What does it take to establish a drug free work place. Are technologies available other than urine testing for pre- employment screening and monitoring of employees. Various methods are now available to screen for illicit drug residues on items handled by individuals. The residues can be acquired from the surfaces of items such as telephones, door knobs, steering wheels, lockers, clothing, identification cards, etc. Test kits are also available for urine testing at NIDA threshold levels. Analysis of hair, saliva, and sweat is now possible. How good ar these methods and kits. What value are they to the public. What are the legal concerns facing employers. What do the screening test show. These questions and others are addressed in this paper. The authors review for the reader how drug abuse by US workers costs businesses. The paper then addresses the various aspects of the DOT regulations to determine why urine analysis (UA) is insufficient to eliminate drug abuse. The authors present applications of screening technologies in addition to UA. Finally, the authors provide a conclusion of findings and recommendations for businesses that truly want or need drug free work places.

A monolayer of antibody on a semimirror comprised of small islands of indium acts as a sensor capable of detecting vapors at extremely low concentrations without the use of wet chemistry. Already shown capable of detecting cocaine vapors at 4 femtograms per cc of air, the use of the device, called an immunoassay film badge, for detecting drugs on the breath is a natural extension of the sensor's use. This paper describes this application and initial experiments that demonstrate its feasibility.

The concept behind a volatile compound mapper, or electronic nose, is to use the combination of multiple gas sensors and pattern recognition techniques to detect and quantify substances in gas mixtures. There are several different kinds of sensors which have been developed during recent years of which the base techniques are conducting polymers, piezo electrical crystals and solid state devices. In this work we have used a combination of gas sensitive field effect devices and semiconducting metal oxides. The most useful pattern recognition routine was found to be ANNs, which is a mathematical approximation of the human neural network. The aim of this work is to evaluate the possibility of using electronic noses in field instruments to detect drugs, arson residues, explosives etc. As a test application we have chosen breath alcohol measurements. There are several reasons for this. Breath samples are a quite complex mixture contains between 200 and 300 substances at trace levels. The alcohol level is low but still possible to handle. There are needs for replacing large and heavy mobile instruments with smaller devices. Current instrumentation is rather sensitive to interfering substances. The work so far has dealt with sampling, how to introduce ethanol and other substances in the breath, correlation measurements between the electronic nose and headspace GC, and how to evaluate the sensor signals.

A walk-through portal sampling module which incorporates active sampling has been developed. The module uses opposing wands which actively brush the subjects exterior clothing to disturb explosive traces. These traces are entrained in an air stream and transported to a High Speed GC- chemiluminescence explosives detection system. This combination provides automatic screening of passengers at rates of 10 per minute. The system exhibits sensitivity and selectivity which equals or betters that available from commercially available manual equipment. The systems has been developed for deployment at border crossings, airports and other security screening points. Detailed results of laboratory tests and airport field trials are reviewed.

An increasingly popular method of transporting modest quantities of narcotics across international borders is to employ 'swallowers'. These are people who typically enter the country as international airline passengers after swallowing small, water-tight packages of heroin and/or cocaine. Rapid and accurate identification of swallowers in the airport environment poses difficult technical changes. Commonly used medical inspection technologies fall into one of two categories. Either they are unsuitable for widespread use, or they do not provide adequate information. An example of the former is x-ray scanning, while an example of the latter is ultrasonic imaging. Quantum Magnetics has developed a system to screen selected airline passengers for the presence of swallowed narcotics. The system utilizes magnetic resonance, which provides the physical basis for the magnetic resonance imaging systems widely used in the medical community as an alternative to x-rays. The system is currently operational, and laboratory performance testing is complete. Both the design of the system and its performance will be discussed. This work was sponsored in part by the Office of National Drug Control Policy and the US Customs Service.

A test and evaluation pilot study was conducted in January 1996 at Sea-Tac International Airport in Seattle, Washington to determine the initial effectiveness of the Millimeter- wave Holographic Weapons Surveillance System. This is a new personnel surveillance systems for the detection of concealed metal, plastic, and ceramic weapons and other threatening materials. Two different frequency bands were used in the study: Ku band and Ka band. Over 7000 Millimeter-wave (MM-wave) holographic images were obtained on 21 different models. The 7000 images were used to produce simulated real-time surveillance system videos. The videos were constructed by obtaining 36 images of the models at 10 degree increments for 360 degree coverage. A library of two hundred videos were produced for this pilot study: 100 at Ku band and 100 at Ka band. The videos contained either a threat or no threat. The threats were concealed at different locations on the models. Various innocuous items and different clothing combinations were also used n the construction of these videos. Twenty-nine certified Sea-Tac screeners were used in the initial test and evaluation of this new surveillance technology. Each screener viewed 160 MM-wave videos: 80 Ku band and 80 Ka band. The ratio of non- threat to threat videos per band was three to one. Test and evaluation software was developed to collect data from the screeners on-line for the type and location of threat detected. The primary measures of screener performance used to evaluate this new technology included, the probability of detection, the probability of a false alarm, measures of screener sensitivity and bias, and threat detection time.

In an effort to avoid detection, smugglers and terrorists are increasingly using the body as a vehicle for transporting illicit drugs, weapons, and explosives. This trend illustrates the natural tendency of traffickers to seek the path of least resistance, as improved interdiction technology and operational effectiveness have been brought to bear on other trafficking avenues such as luggage, cargo, and parcels. In response, improved technology for human inspection is being developed using a variety of techniques. ASE's BodySearch X-ray Inspection Systems uses backscatter x-ray imaging of the human body to quickly, safely, and effectively screen for drugs, weapons, and explosives concealed on the body. This paper reviews the law enforcement and social issues involved in human inspections, and briefly describes the ASE BodySearch systems. Operator training, x-ray image interpretation, and maximizing systems effectiveness are also discussed. Finally, data collected from operation of the BodySearch system in the field is presented, and new law enforcement initiatives which have come about due to recent events are reviewed.

Many different techniques have been investigated for detecting weapons, explosives, and contraband concealed under a person's clothing. Most of these are based on imaging the concealed object by using some sort of penetrating radiation, such as microwaves, ultrasound or electromagnetic fields.In spite of this effort by dozens of research groups, the only technique that has resulted in a commercially viable product is back-scatter x-ray imaging, as embodied in the SECURE 1000 personnel screening systems. The SECURE technology uses radiation levels that are insignificant compared to natural background values, being viewed as 'trivial' and 'completely insignificant' under established radiation safety standards. In the five years since the SECURE 1000 was developed, more than a dozen field trials and initial placements have been completed. This paper describes both the capabilities and limitations of the technology in these real-world applications.

Automated Fingerprint Identification has a history of more than 20 years. In the last 5 years, there has been an explosion of technologies that have dramatically changed the face of AFIS. Few other engineering and science fields offer such a widespread use of technology as does computerized fingerprint recognition. Optics, computer vision, computer graphics, artificial intelligence, artificial neural networks, parallel processing, distributed client server applications, fault tolerant computing, scaleable architectures, local and wide area networking, mass storage, databases, are a few of the fields that have made quantum leaps in recent years. All of these improvements have a dramatic effect on the size, speed, and accuracy of automated fingerprint identification systems. ThIs paper offers a historical overview of these trends and discuss the state of the art. It culminates with an overview an educated forecast on future systems, especially those 'real time' systems for use in area of law enforcement and civil/commercial applications.

Drug abuse is highly correlated with criminal behavior. The typical drug-using criminal commits hundreds of crimes per year. The crime rate cannot be significantly reduced without a reduction in the percentage of the population abusing drugs and alcohol. Accurate and timely estimation of that percentage is important for policy decisions concerning crime control, public health measures, allocation of intervention resources for prevention and treatment, projections of criminal justice needs, and the evaluation of policy effectiveness. Such estimation is particularly difficult because self reporting is unreliable; and physical testing has to date required blood or urine analysis which is expensive and invasive, with the result that too few people are tested. MIKOS Ltd. has developed a non-contact, passive technique with the potential for automatic, real- time screening for drug and alcohol use. The system utilizes thermal radiation which is spontaneously and continuously emitted by the human body. Facial thermal patterns and changes in patterns are correlated with standardized effects of specific drugs and alcohol. A portable system incorporating the collection and analysis technique can be used episodically to collect data for estimating drug and alcohol use by general unknown populations such as crowds at airports, or it can be used for repetitive routine screening of specific known groups such as airline pilots, military personnel, school children, or persons on probation or parole.

One method to enhance or analyze fingerprint images is to use a directionally applied enhancement operator. Critical to this operator is the correct local direction. Typically, the local direction is found by correlating lines of constant extent but different angles with the local gray level intensity. The extent of these directions is a trade- off between performance in high-curvature and noisy regions. If the extent is too short the correlation works well in high-curvature regions but fails in noisy ones. The longer extent does well in noisy regions but fails in high- curvature ones. This paper shows how the directional image enhancement can also be done in the frequency domain. This approach overcomes the noise and high-curvature limitation of the spatial domain methods.

A non-minutiae based technique for fingerprint matching is presented, based upon spatial correlation of fingerprint grayscale images. Normally, such techniques suffer from plastic distortion of the fingerprint ridges, which varies irreproducibly from one scan to another. This distortion can be seen as a process which reduces the degree of global spatial coherence of the fingerprint images. In the proposed approach, this limitation is accommodated by correlating small, windowed regions of a test image with the entire reference image. A score can then be ascertained, based on the partial correlations of these regions. The technique lends itself naturally to implementation by optical means.

This paper presents a new fingerprint sensor for automated fingerprint identification systems. This fingerprint sensor consists of alight source, a charge coupled device and fiber optic faceplates. Because of the fiber optic faceplates, this fingerprint sensor does not have the space for image formation that a lens has and thus can be miniaturized. The size of the prototype fingerprint sensor, which has a 12 mm by 18 mm fingerprint input face, is 48 mm by 72 mm by 44 mm. The verification performance is evaluated by using this prototype.

Recent advances in image and pattern recognition technology- -especially face recognition--are leading to the development of a new generation of information systems of great value to the law enforcement community. With these systems it is now possible to pool and manage vast amounts of biometric intelligence such as face and finger print records and conduct computerized searches on them. We review one of the enabling technologies underlying these systems: the FaceIt face recognition engine; and discuss three applications that illustrate its benefits as a problem-solving technology and an efficient and cost effective investigative tool.

One of the common techniques of printing gray scale images using a bi-level device such as a commercial laser printer is Error Diffusion. Since most of the fingerprint images are gray scale images, they too can be printed using error diffusion. However, none of the existing popular error diffusion algorithms makes use of the ridge-flow information present in a fingerprint image. In this note, we explore the use of ridge-flow directions in a fingerprint image for the purpose of the error diffusion in printing fingerprint images.