12 July 1997
From Office of Law Enforcement Technology Commercialization (OLETC)
Argonne has developed, under DOE Arms Control funding, a unique method for high-reliability item identification by identifying microscopic surface "fingerprints". This is done by comparing digital scanning electron microscope (SEM) images of surface replicas taken from an inspected item. Some commercial applications of this include identification of part forgeries and rare coins. Another related application is surface inspection of parts that must remain in use. The advantages of this surface identification inspection technique are that the plastic replicas are very inexpensive, can be readily made in the field, and are readily transported to the SEM lab, and the tagged or inspected part can remain in the field.
Digital Image Comparison for Security
Argonne has extensive experience with software/hardware image processing for Arms Control work in which digital images are compared for similarity. This work led to a demonstration of a security system based upon changes in a video digital image. The unique advantage of this system is that only the regions in an image that would change upon tampering are examined for changes.
HIDEW Software for Remote Sensing
Argonne researchers have developed the Hyperspectral Imagery Data Exploration Workbench (HIDEW), a software tool incorporating sophisticated computer techniques to analyze digital imagery for remote sensing purposes. HIDEW is currently being used to identify terrestrial ecosystems and evaluate mechanisms such as the effect of climate change on biota and the corresponding biotic feedback to the climate. It could just as easily be applied to law enforcement tasks that need to identify images or parts of images, or search for patterns in digital images. The system allows scientists to recognize visual textures and infer physical properties that would otherwise be indistinguishable.
HIDEW has been developed to enable analysts to conveniently access and manipulate high spectral-resolution imagery. The considerable size of this type of data set poses a significant management challenge. HIDEW provides sophisticated analysis tools accessed by a modern user interface to allow intuitive operation by its users. It is also suitable for extracting specific data from large data sets to feed other computer analysis tools with rigid data input requirements that cannot deal with the very large original data sets.
Hyperspectral Pattern Matching
This work will exploit hyperspectral imagery analysis technology developed by ANL. An image of a human face is scanned using a high-quality CCD camera. The image is then reduced to a fractal file in a database. At appropriate or necessary times, a group of persons entering or leaving a control point can be scanned, and the results compared with the database. Any face not previously scanned will be identified.
Argonne proposed a system to the U.S. Customs Service last year that would provide a means to detect and characterize very-low-level airborne, and surface marine, traffic to and from such sites as unattended off-shore platforms. Utilizing a large number of inexpensive sensors in "suites" with some signal recognition capabilities, and linked via very fine fiber-optic cable, one can create an intelligent network that can identify, characterize, and record acoustic and other signals from vehicles passing within a specified distance of a platform. These signals can then be collected and analyzed to determine whether a particular helicopter or vessel followed a particular route, or visited certain platforms. This system would leverage work done for the ARIADNE project for the U.S. Navy in detection of underwater objects. It also would utilize an extensive collection of existing research on the identification and characterization of aircraft acoustic signatures.
Data Access & Analysis
Argonne has done extensive work for the Joint Chiefs of Staff, US SOCOM, DoE, and others in developing systems to seamlessly access multimedia information from selected distributed databases and retrieving, analyzing, and archiving information of interest to intelligence analysts. These tools have obvious application in law enforcement
Lessons Learned Systems
Argonne has developed an automated system for accessing lessons learned to support military planning. Lessons learned information could include after-action reports, training exercise results, planning heuristics, subject-matter expertise in a key problem domain, demonstrated solutions to operational problems, and other information. The lessons learned system accesses records in a number of predetermined formats from other automated systems as well as from records entered directly into the system. Although each lessons learned system already has its own structured format, an interface is provided so that the systems are transparent to the user. The system, which can access lessons learned developed in different formats and residing in different databases, also has applicability to nondefense areas, such as environmental management and energy facility operation. In environmental cleanup, experiences and miscellaneous bits of information with approaches to dealing with specific wastes could be valuable as the cleanup process (analogous to the military planning process) moved forward. A lessons-learned system could also be applicable to facility operation if a problem or situation developed for which previous solutions has been recorded.
Access to Databases
Argonne has developed the Application Interface Engine (AIE), an object-oriented computer software development environment that is being used by ANL, government organizations, and industry to develop flexible and portable software applications. The AIE follows open architecture standards, is nonproprietary, and is subject to a comprehensive QA/CM process. It is available on both the Sun Workstation (X Windows using either OPEN LOOK or Motif) and on PCS (Microsoft Windows). This means that the same application will run on both PC and Sun platforms transparently by linking with the appropriate AIE libraries. This feature greatly decreases the burden of maintaining multiple versions of the application. One of the strengths of the AIE is that it allows various parts of an application to be located on different machines, and allows access to different databases (using different database management systems) that are located on different machines.
Information Systems for Law
Large-scale information retrieval systems to manage and search volumes of text information are cornerstone components to law enforcement agencies which compile vast amounts of crime information. This information consists of free text, graphics, photos, images, and possibly audio and video. However, unless these vast amounts of information can be efficiently and timely searched for similar crime cases and distinguishing characteristics, law enforcement activities are impeded. With the emergence of new technologies in information retrieval and analysis, these large databases can be collected, searched, and analyzed by agents with the application of advanced text retrieval tools. These technologies include concept searches, boolean searches, natural language searches and hyperlinks to text, graphics and multimedia components. All data on a case can be organized together through this hyperlink technology in a central repository. Various artificial intelligence tools can be effectively utilized to provide intelligent assistance to agents analyzing and locating information.
The Advanced Computer Application Group of Argonne National Laboratory has extensive experience in full text information retrieval systems on PCS, local area networks and Internet through the World Wide Web (WWW) and Wide Area Information Server (WAIS) servers, leveraging off the expertise of production systems developed for the Department of Energy. Data can be effectively interchanged over secured local area networks and Internet connections to allow geographically separated departments to rapidly exchange data and search these independently maintained databases of information.
Detection of Narcotics and Other
Argonne is currently engaged in research and infrastructure support for the Counterdrug Technology Assessment Center of the Office of National Drug Control Policy (ONDCP). CTAC is responsible for coordinating and overseeing all federally funded research in the counter-narcotics area. There are more than 50 federal departments, agencies, and bureaus engaged in drug control activities. Argonne's current activities include the evaluation of nuclear, x-ray, and chemical technologies for detection of contraband (narcotics, explosives, nuclear materials, etc.) and the support of federal, state, and local law enforcement agencies (LEAs) in increasing the use and effectiveness of advanced technologies. Argonne has worked with the U.S. Customs Service, the U.S. Coast Guard, ARPA, FBA, DEA, BATF, and others.
Argonne has developed and coordinated field tests at Drug Enforcement Administration sites to evaluate instrumentation used for the detection of illicit drug laboratories. Technologies were developed to reproducibly release both large and small quantities of analytes. Argonne also has been involved in the passive-remote detection of illicit substances using Fourier Transform Infrared (FTIR) spectroscopy. The objective is to track vapors from moving platforms such as helicopters or cars. Miniature equipment also was developed and tested for use in covert activities. In addition, the Laboratory has a program with the Federal Aviation Administration to evaluate nuclear techniques for the detection of explosives in luggage.
Argonne has demonstrated the ability to detect cocaine inside a sealed container, using a nontoxic simulant that was an accurate mockup of the chemical element composition of cocaine hydrochloride. This demonstration, performed in response to a suggestion by the DOE program sponsor (NN-20), used the Argonne "APSTNG" system, which interrogates a volume to be inspected by a low-dose neutron source, and detects the characteristic elemental gamma-ray spectra. Using a flight-time technique the "APSTNG" can provide a coarse tomographic image of the distribution of detected elements, behind walls or inside sealed containers. Explosive and other contraband can also be detected.
Safeguards and Security Training/Orientation
The Information and Publishing Division has the following capabilities that may be relevant to law enforcement. The Video Group in IPD-Media Services has produced videos for use by DOE and ANL Safeguards and Security, and the creation of videos on law enforcement topics for other groups is a current IPD capability. Additionally, IPD provides many information services (graphic design, editing, writing, information retrieval and organization, and computer security) that could be extended to law enforcement groups.
Human Face Recognition
Argonne proposes to develop a neural-network-based face recognition system for law enforcement applications. The system will be an adaptation and integration of neural network and computer vision technology already developed at Argonne for such applications as 3D cardiac motion analysis, 2D neural imaging systems for automated crystal inspection, and automated weld certification. The research draws on ANL's experience with neural networks, fuzzy logic, and advanced pattern matching innovations.
Advanced Capabilities in Trace Elemental
Trace elemental analysis ("fingerprinting") of evidence is an extremely powerful tool in forensic investigations. Argonne National Laboratory has one of the best equipped and staffed analytical chemistry laboratories in the United States for trace elemental analysis. Relevant analytical capabilities in the ACL include scanning electron microscopy (SEM), X-ray diffraction (XRD) and fluorescence (XRF) spectrometry, furnace and flame atomic absorption (AA) spectrophotometry, furnace and flame atomic absorption (AA) spectrophotometry, inductively coupled plasma mass spectroscopy (ICP-MS). Argonne has world-class expertise in the application of these techniques, and its staff also are also highly experienced in the preparation of small and/or one-of-a-kind samples for the determination of major, minor, and trace elemental composition. In addition, Argonne has a formal quality assurance program that includes sample chain-of-custody and analysis quality control procedures to ensure sample integrity and analysis data quality.
Advanced Organic Analysis
To determine organic compounds e.g. solvents, illicit substances, and explosives at trace levels, Argonne National Laboratory has the following capabilities; liquid chromatography/mass spectrometry, gas chromatography/mass spectrometry, supercritical fluid extraction and chromatography, Fourier transform infrared (far, mid, and near) and Raman spectroscopy. These techniques can be used to analyze volatile, semivolatile and heat-sensitive compounds at parts-per-trillion level. Some of these capabilities have been used to determine an illicit substance on U.S. paper currency at trace levels.
Forensic Laboratory Performance Evaluation
Laboratory analyses or tests provide key information in many court cases; confidence in those results can be a critical factor in determining the outcome of a trial. One major indicator of confidence is a laboratory's participation and successful performance in external proficiency testing (PT) or performance evaluation (PE) programs for those analysis types or tests the laboratory conducts. The Analytical Chemistry Laboratory (ACL) at Argonne National Laboratory (ANL) has technical expertise in the use and interpretation of PE program results; and has developed for the DOE a centralized, integrated system for the assessment of environmental laboratory performance in PE programs. The system includes the collection of PE study data from multiple programs, assessment of laboratory performance using those data, and subsequent distribution of that information for use in evaluation of laboratory performance. Integration of results from several PE programs allows enhanced use of the data from individual PE programs. Centralization provides consistency of data interpretation, and eliminates the need for redundant assessments by various end-user organizations. This can be particularly important in the case of small organizations with fewer qualified personnel. In addition, the centralized system can provide a list of qualified laboratories, so organizations needing laboratory analysis services can find a laboratory with acceptable performance to do their work. Distribution of PE and PT program data via Internet makes the information easily accessible. PE and PT Programs are applicable to forensic laboratories and this integrated, centralized approach could greatly enhance the use of these programs to the law enforcement community.
Intalligent Gas Sensor
Only 1 mm Square, a miniature gas sensor can detect multiple gases at very low (ppm) levels. The sensor can operate from battery sources and is suitable for gas detection instrument development. Potential applications for the sensor include DEA, vehicle inspection, and explosives detection.
Using Hyperspectral Image Processing
for Remote Reconnaissance
Using neural processing algorithms and parallel computing techniques, Argonne has developed a very quick method for extracting useful information from hyperspectral data developed from photos. Hyperspectral imagery allows for a very fine discrimination and definition of objects and their characteristics at remote locations. DEA could use this method to identify illegal crops or emissions from drug manufacture. The method may also be used to track motion (vehicles, ships, etc.) based on temporal changes in complex hyperspectral signatures. Hyperspectral imaging may also be applied to monitoring suspects, checking them for trace elements, changes in body chemistry, and contaminants. It might also be used for reconnaissance in remote locations, such as assessing vandalism at southwestern U.S. archaeological sites.
Expert Systems Design for Security
Expert systems can provide decision support and map history of criminal activities in an area. It can also develop MOs for criminals and areas, using this information to predict crimes before they happen. The system can also identify suspects based on heuristics and direct officers to overlook phenomenon; can look for patterns in alibis and detect irregularities in statements, and can be used to verify statements.
Cocaine Transfer From Currency to
Research has shown that the cocaine contaminating much of the paper money circulating in major cities cannot easily rub off on the hands of people who handle it. Cocaine becomes imbedded in the currency fibers and does not rub off easily on the fingers of people who handle the currency.
For further information, contact:
Industrial Technology Development Center
Argonne National Laboratory
9700 S. Cass Ave., Bldg. 900
Argonne, IL 60439
1-800-627-2596, FAX 708/252-5230
Click here for additional information on Argonne capabilities
Last Modified: Monday, April 14 1997 09:59