Interview with an Expert

How the NIJ’s AFIS Working Group
will affect the future of AFIS interoperability

An exclusive Evidence Technology Magazine interview with
John S. Morgan, Ph.D.
Deputy Director for Science and Technology
National Institute of Justice

There have been major milestones in the history of fingerprint identification. In 1924, the nucleus of the FBI fingerprint paper file was formed. By 1971, that file had grown to more than 200 million cards. In 1977, the very first local automated fingerprint identification system (AFIS) began its operation. The year 1999 brought the FBI’s Integrated automated fingerprint identification system (IAFIS). Today, law enforcement is looking forward to yet another major milestone: the efficient interoperability between local, state, and national identification systems. To learn more about AFIS interoperability, we spoke with Dr. John S. Morgan, Deputy Director for Science and Technology at the National Institute of Justice (NIJ).

EVIDENCE TECHNOLOGY: There are a lot of fingerprint specialists out there who will be very glad to see improvements in the interoperability between different AFIS systems.
MORGAN: I’m sure they will be glad because there have been some issues. If you look at the technology of DNA—which is a much newer form of identification than fingerprints—the interstate interoperability problem never occurred. The national DNA index system—CODIS—provides perfect interoperability among all the DNA laboratories in the country. They all collect the same 13 STR markers. And they are all able to make matches with other states through the national index. As a result, we routinely see DNA hits that occur outside the home jurisdictions where the cases arose. Unfortunately, that sort of thing does not occur nearly as often as it should with fingerprints. We need to make sure that fingerprints, like DNA, can take advantage of interoperability seamlessly and have a much bigger impact as a result.

EVIDENCE TECHNOLOGY: Please fill us in on what we can expect in the future with AFIS interoperability.
MORGAN: We should probably start with AFIS systems themselves. Auto-mated fingerprint identification systems allow forensic practitioners to identify individuals based on friction-ridge images. AFIS systems are important automated tools that can assist the latent-print examiners and speed their ability to determine who would be the most likely match against a particular latent image. I think many readers of Evidence Technology Magazine can remember the days before automated fingerprint identification systems, when one would search latent prints against ten-print cards manually. That process was almost impossible without a relatively short list of suspects who had fingerprints on file.

EVIDENCE TECHNOLOGY: Some will remember. Maybe not all…
MORGAN: AFIS systems, of course, let you do what you used to have to do with paper systems, but with much more efficiency. They are a critical element in improving the work of the fingerprint examiner in state and local law enforcement. AFIS systems certainly represent a revolutionary improvement in their capability. And that’s why the Department of Justice (DOJ) has actually spent a fair amount of resources improving AFIS technology at the state and local levels. They are still continuing to offer grant programs to do this.

EVIDENCE TECHNOLOGY: What kind of programs do you refer to?
MORGAN: Well, the National Criminal History Improvement Program—or NCHIP—actually contributes a fair amount to the purchase of AFIS technology. NCHIP is a program of the Office of Justice Programs. AFIS systems can also be purchased with funds from Justice Assistance Grants. And NIJ is sponsoring research to improve the system, such as our Fast Capture Initiative. A lot of people use their criminal-justice formula grants to buy these systems. Generally speaking, there is a significant yearly investment by the DOJ for improvements in the AFIS systems.

EVIDENCE TECHNOLOGY: How does funding apply to our discussion?
MORGAN: The issue of funding comes up because of interoperability. Inter-operability is not just a problem that applies only to AFIS systems. It cuts across a lot of different disciplines in law-enforcement technology. For example: There are interoperability problems in law-enforcement radio systems. There are interoperability problems in just about every informational system that’s out there. And there are obviously some very significant interoperability problems in AFIS systems, as well.

MORGAN: There are several manufacturers of AFIS systems in the United States that are more or less actively marketing to law-enforcement agencies. Some of them will reference common standards—but in general, they each use a proprietary technology and proprietary data-storage and com-munication standards. Some of that is unavoidable, of course, because manufacturers have different approaches to how they do their print matching. Different manufacturers use very different algorithms to break apart fingerprint information. Because the underlying algorithms look at fingerprints differently, the systems will store the data differently. And as a result, they will communicate differently, as well.

MORGAN: To some extent, it was not really unexpected that interoperability problems would occur as the market evolved. But, nevertheless, we believe that it is possible for the systems to talk to one another using some sort of common format, even if they have an underlying proprietary technology.

EVIDENCE TECHNOLOGY: How do you get these propriety formats to work together?
MORGAN: In some areas, problems have been solved through cooperative efforts. Why? Because the issues that arose were really policy based—even though they were to some extent caused by technology. State and local agencies—and the federal government, as well—need to do more cooperating with each other in order to ensure that the interoperability has been addressed with respect to AFIS systems and other systems that also have inter-operability problems.

EVIDENCE TECHNOLOGY: Has anything substantive been done yet?
MORGAN: Yes. The Western Identifi-cation Network (WIN) has very good interoperability. They are using what amounts to a single system for doing their AFIS work. That has been very successful in improving interoperability problems among the states involved. So there is some real potential with respect to improvements of interoperability, although at this point there is still no national framework for solving the AFIS interoperability problem.

EVIDENCE TECHNOLOGY: Is there any significant movement toward solving the problem on a national scale?
MORGAN: Yes, there is. First, I am going to give credit to the FBI’s Criminal Justice Information Services (CJIS). They have developed a basis for moving ahead with national standards. And we here at NIJ believe that using the CJIS standards is a very good framework for achieving interoperability. If everyone would adopt their data standard, I think we would solve many interoperability problems. And in order to try to bring everyone together on this and encourage interoperability, NIJ is establishing the AFIS Working Group. In addition to the manufacturers of AFIS systems, the group will include representatives from NIJ, from the National Institute of Standards and Technology (NIST), from the FBI, and from the state and local practitioner community. What we hope to do is bring together the researchers—such as those from NIST—and the user community, the system vendors, and the leading criminal-justice associations to develop a roadmap for AFIS interoperability.

EVIDENCE TECHNOLOGY: And where are you now on that working group?
MORGAN: We have identified the organizations and the individuals and we have spoken with the people who will be part of the group. The AFIS Working Group will meet in early 2008 and focus on technical and policy issues that are barriers to interoperabilty.

EVIDENCE TECHNOLOGY: What else is happening?
MORGAN: There are other groups that are looking at this, as well. There is an general assessment being made of the forensic sciences by the National Academy of Sciences (NAS). They’re doing an overall look, under congressional direction, at forensic sciences and the practice of forensic science. One of the tasks given to the NAS by Congress was to look at AFIS interoperability. They will be reporting a broad set of recommendations concerning forensic science. We expect that report sometime in the spring, although they are on their own timeframe. But they will be addressing the AFIS interoperability issue, as well.

EVIDENCE TECHNOLOGY: Anything else regarding interoperability?
MORGAN: Yes. NIJ is doing other things to address the interoperability problem. As I indicated, there is an issue of policy at the state and local level with respect to interoperability. One policy issue has to do with being willing to pay a little bit more for an AFIS system if it is interoperable with other states and with national standards. Part of the task is proving that interoperability is valuable, as well as identifying the challenges to achieving interoperability. For example, we know that departments are worried about liability and security issues when other agencies access their data systems. To approach this, NIJ has a contract with the International Association for Identification (IAI) to look at the impact of AFIS interoperability between the states of New York and New Jersey. What we will be doing is examining how many latent-print identifications are made in those states under four conditions. First: If you just don’t use an AFIS system. Second: If you use an in-house AFIS system but without the interoperability of another AFIS system. Third: How it works on an interstate basis, between New York and New Jersey. And fourth: How well it works utilizing the FBI’s fingerprint database, IAFIS, as a way to do print identification.

EVIDENCE TECHNOLOGY: What do you hope to learn with this study?
MORGAN: What we hope will come out of the study is information about the impact—the positive impact—on latent-print matching when you have interoperability outside your own jurisdiction. When we are able to demonstrate to police chiefs and to departmental policy makers that AFIS interoperability is very important for capturing criminals, then it will be relatively easy for them to say, “Yes, it’s worth it to invest in systems that are interoperable.”

EVIDENCE TECHNOLOGY: So this is really a key issue in law enforcement…
MORGAN: Yes, it is very important. By bringing the vendors to the table in the AFIS Working Group and convincing law-enforcement leaders that this is a key issue, we believe that the prospects are quite good for solving the AFIS interoperability problem in some reasonable time frame. At the very least, we should be able to provide a framework under which the problem can be solved.

EVIDENCE TECHNOLOGY: Isn’t there a problem with upgrading systems?
MORGAN: Yes, there are issues with respect to upgrading out there. And I think that a lot of state and local agencies would like to see upgrades in their equipment regardless of the interoperability issue. But that’s not exactly part of what we’re looking at right now at NIJ. Right now, our focus is on interoperability, per se.

EVIDENCE TECHNOLOGY: Are other changes going to need to happen?
MORGAN: I do know that CJIS has been very concerned about handling large volumes of searches. They feel like it would be an extraordinary challenge to handle a huge amount of interstate traffic on latent-print match-ing. It’s difficult for them to handle even the current traffic of tenprint matching. It is a big challenge and they’ve been doing a remarkably good job. But latent-print matching is more difficult and would require a big investment. Part of the issue here—because you have many different systems that have many different algorithms—is how good are the different systems in the matches they make and how much capacity do they have if they start to build up the inquiries that are made to the state-to state systems? Now, we do have a research program under way that we hope will improve the speed and the quality of tenprint collections using electronic means.

EVIDENCE TECHNOLOGY: Can you provide some details about that?
MORGAN: We have a research-and-development program on Fast Capture of fingerprints and palmprints that we have been working on for a couple of years now with joint funding from the Department of Justice, the Department of Homeland Security, and the Department of Defense, with cooperation from the Department of State. All of them are very interested in doing a better job of collecting tenprint and palmprint image data. We have technology developments in progress that we hope will give us the ability to collect tenprints directly to electronic files in less than 15 seconds—and the ability to pull palmar impressions in less than a minute. Palmprints, incidentally, are very important, since—depending on who you ask—something like a quarter or more of identifiable latent prints are actually from the palmar surface as opposed to fingers. We hope that this development effort will speed the collection of tenprint data. And we also hope that it will improve the overall quality of data in AFIS systems and in IAFIS. That is something of a big deal in this case, because being able to reliably collect extremely high-quality database images is going to make latent-print searching much more effective. You must have excellent image data to match against in order for all of these systems to work well. That’s a long answer. The short form is this: We are really hoping that we’ll be able to use the Fast Capture Initiative and the available technology to upgrade the ability of law enforcement for that kind of work.

EVIDENCE TECHNOLOGY: Are you at liberty to tell us what stage of development the Fast Capture Initiative is in at this time?
MORGAN: We have some prototype technology that has been delivered to us. We are going to be doing an evaluation of both the technology developed under our program as well as some commercially available technology that is based on a slightly different approach. We are going to be doing those evaluations at NIST starting this year. The technology we have prototypes on is extraordinarily good and very fast, and I’m hopeful that they’ll do well when we do the evaluation at NIST. However, I think that there is going to be another generation of systems in the 2009 or 2010 timeframe—and those will probably give us a very strong set of electronic-scanning equipment. We will just have to see what the 2008 NIST evaluation shows. In the meantime, we have
prototypes right now that are amazing. The level of detail that they can see—sometimes referred to as Level 3—is very, very good, and is well beyond anything that you would be able to collect using the conventional rolled-ink methods of tenprinting. And these systems are so fast…

EVIDENCE TECHNOLOGY: So this is technology we haven’t seen before?
MORGAN: Right. The one system that I’m really excited about is capable of taking an image of your sweat pores actually sweating. It can not only show an individual sweat pore, but it can image the sweat coming out of the pore and evaporating—all in real time. It is obviously a very fast and very detailed technology in order to be able to come up with that.

EVIDENCE TECHNOLOGY: With regard to AFIS interoperability, is there anything our readers who are at the state and local levels can do to move things along more quickly?
MORGAN: Actually, they are the most important part of this. The state and local agencies can move the vendor community very quickly to interoperability by demanding that particular quality in their procurements. You have to realize that, even though the up-front costs may be slightly higher for systems that include interoperable standards, the long-term benefits of such systems are clearly worth the added expense. That’s why I strongly encourage state and local users to require interoperability in their AFIS procurements. They have already had an impact. Because of the actions of state and local users, the manufacturing community is willing to join the AFIS Working Group and help find a long-term solution to interoperability.

Are vendors going to have to come up with some sort of converters?
MORGAN: Many of the vendors have already done that. There are already systems that can communicate in a standardized format with other vendors’ systems. Even though their proprietary algorithms are protected, their systems are still basically interoperable. That is something the vendor community has already had to face—and it is some-thing that it may need to do some more work on in coming months or years. But it is not difficult work. After all, it’s not a technical problem. It’s a policy problem.

EVIDENCE TECHNOLOGY: Do we get into a capacity problem when we go from one local system to another?
MORGAN: Well, it’s both a capacity problem and a cooperation problem. If there is an incident that occurs on an Interstate highway in…let’s say Kansas…and the perpetrator comes from Indiana, then they need to go today to Indiana and look for the suspect in Indiana’s AFIS system. And if such searches were automated, Indiana would have to provide enough capacity to deal with the increase in the activity in their AFIS system. This could be an issue in the future. Some of it will involve policies and procedures. Obviously, home jurisdictions are going to have to routinely look within their own AFIS system first and not ask to do broad interstate searches without having looked in their home systems. They might have to prioritize interstate matching based on some understanding of the likelihood of where the most likely hits would occur. A lot of that kind of work is already being done out there using paper and fax machines instead of computers. So issues like that are not going to be new to the latent-print community.

EVIDENCE TECHNOLOGY: And all of these issues will be addressed during the year by the AFIS Working Group, is that right?
MORGAN: Yes. We’re going to look at the broad range of issues. Our primary purpose is going to be to provide a framework for AFIS interoperability. Of course, in terms of more structural issues in the latent-print community, the AFIS Working Group can only go so far. We will not be in a position to serve as advocates for funding more AFIS systems around the country and things like that. But we will be taking a look at that—and all of the other potential problems or issues—so that we will have an understanding about what the community would need to do with respect to policy and procedure as AFIS interoperability improves.

EVIDENCE TECHNOLOGY: The AFIS Working Group will be highlighting a list of specific needs. But will it also be making specific recommendations to address those needs?
MORGAN: We will be making some recommendations, but we will not be making any recommendations that involve federal funding. That is not our role in this effort.

EVIDENCE TECHNOLOGY: If you were sitting here talking with a latent-print examiner from a local law-enforcement agency—or a chief of police or public administrator, for that matter—what would you tell that person about the future of AFIS interoperability?
MORGAN: I would tell them that they are the ones that will initiate change. The fingerprint specialists at the local level will really play a key role in demanding AFIS interoperability from the vendor community. Of course, law -enforcement executives must take the leadership role to solve the policy issues surrounding interstate access and interoperability. Quite a bit has been accomplished already.

EVIDENCE TECHNOLOGY: What to you mean? In what way?
MORGAN: Well, the manufacturers and vendors of AFIS systems are much more open to a national structure for interoperable standards than they were just a couple of years ago. I would also tell them that there is light at the end of the tunnel. The federal government is working in many ways to organize the field so that we will be able to achieve that goal.

EVIDENCE TECHNOLOGY: Thank you for speaking with us today.

A brief look at the background and experience
of John S. Morgan

John S. Morgan, Ph.D. is Deputy Director for Science and Technology at the National Institute of Justice (NIJ). He manages the agency’s science-and-technology portfolios and provides strategic-science advice for the Director and the Department of Justice. Dr. Morgan directs a wide range of technology programs for criminal justice, including DNA, less-lethal technologies, body armor, information technology, and communications. In 2007, Dr. Morgan and his team of researchers, lawyers, and analysts received the Service to America Medal in Justice and Law Enforcement for their work with the President’s DNA Initiative. Their efforts have helped solve thousands of cold cases and have dramatically expanded the capacity of local law enfocement to use DNA evidence. Prior to coming to the NIJ, he conducted research in the detection and mitigation of weapons of mass destruction at the Johns Hopkins University Applied Physics Laboratory. He developed mass spectrometry systems for detection of chemical and biological warfare agents; studied methods to protect aircraft from terrorist attack; and developed building and infrastructure protection strategies. His research interests have also included non-destructive evaluation, spacecraft contamination control, high-temperature superconductivity, and high bandgap semiconductors. Dr. Morgan received his Ph.D. in Materials Science and Engineering from Johns Hopkins University in 1990.


January-February 2008 (Volume 6, Number 1)
Evidence Technology Magazine
Buy Back Issue

< Prev   Next >

Forensic Podiatry (Part Two of Two)

THE DISCIPLINE of forensic podiatry—or, in other words, the examination of pedal evidence—has progressed significantly over the past ten years. It is no longer a question of “What can you do with a footprint?” but rather, “Who can we use to evaluate the footprint?” Cases involving pedal evidence, especially bloody footprints and issues of determining shoe sizing or fit issues compared to questioned footwear, have become more common over the past two or three years.