The country’s sense of security vanished in the flames and smoke of that September morning.

To counter America’s loss of security, to reduce its vulnerability, America is depending on its scientists, including a significant number of Indian Americans, to come up with better solutions and better technology. One such solution is the use of Biometrics, the science of using digital technology to identify individuals based on unique physical and/or biological qualities.

Which is why more than twenty Indian American scientists in academia and industry are leading these efforts in biometrics.

Simply put, biometrics is the ability to identify a person through the use of technology and biology.

There are many types of biometrics, but among the most common are scanning fingerprints, voices, faces, retinas or irises. Computer hardware and software programs have been developed to scan a thumb print, for example, and then compare it with a stored databank of other prints for an exact match. Or a voice is compared to a bank of voice samples using algorithms. Face recognition is the measurement of certain characteristics, such as the distance between eyes. Retina scanning has the computer camera inspecting the pattern of veins in a human eye. And, finally, iris scanning takes retina scanning one step further by concentrating on the color pattern surrounding one´s pupils.

In the end, it all boils down to one question: How do we know if someone is who he says he is?

To be sure, biometric technology has its share of critics. Human rights and civil liberties groups have complained that the storage of such information could be subject to abuse and could be used to record people´s movements. Experts have also questioned the efficiency of the technology and said it could leave millions of people vulnerable to a mistaken identity.

Here are some of the most prominent Indian Americans active in biometrics research.

Anil K. Jain

At Michigan State University, the Pattern Recognition and Imaging Processing Research Lab in the Department of Computer Science and Engineering has been working to improve the accuracy of recognition systems such as facial profiling and iris and fingerprint recognition.

Dr. Anil Jain, distinguished professor of computer science and engineering, who directs the laboratory, asserts that people “are viewing biometrics as a solution to the problem of catching and identifying terrorists.” Writing in Communication of the ACM, Professor Jain says: "Biometrics provides a better solution for the increased security requirements of our information society than traditional identification methods such as passwords and PIN.”

For example, says Jain, a person’s face is photographed and the image is compared with those in the database. If there is a “hit,” the person may be stopped, interviewed or searched. If the image does not match any found in the database, it is discarded.

The real problem, according to Jain, is that the technology isn’t very accurate. “There’s a certain amount of hype associated with biometrics. After Sept. 11, vendors began saying that if we’d had these systems in place at the time, we could have avoided the attacks, but I don’t believe that’s true,” he says. “The accuracy can be affected by lighting conditions, by makeup or by facial hair.”

Fingerprint and iris recognition systems are much more accurate but, unlike facial imaging, require the cooperation of the person being identified. “Nothing easy or non-invasive exists at this time that will help us identify terrorists,” acknowledges Jain, who recently gave a presentation entitled, “Biometric Authentication: How Do I Know Who You Are?” during which he addressed broad issues surrounding biometrics and presented various applications of biometrics.

Professor Jain can usually give a rundown on such seamy practices as impersonation and fraud as if they were everyday occurrences. And statistics say he´s right. Billions of dollars are lost each year from the fraudulent use of welfare benefits, credit cards, cellular phones, and bank accounts. Moreover, with people increasingly choosing to buy and sell everything from automobiles to stocks and bonds over the Internet, the gate has been opened wider still for abuses to occur.

Jain, who has co-edited a book titled ‘Biometrics: Personal Identification in Networked Society’ with Ruud Bolle and Sharath Pankranti of the IBM T. J. Watson Research Center, describes biometrics as the measurement of certain physical, physiological, or behavioral characteristics that exhibit the following properties: universality (everyone has one); uniqueness (it is different for different people); permanence (it doesn’t change much over time); and collectability (it’s quantifiable).

That’s why our voices, facial features, fingerprints, eyes (iris and retina), ears, hands, body heat emissions (thermograms), DNA, and smell, and the way we walk, strike a keyboard, and sign our name are all being studied today as potentially useful biometrics tools. However, admits Jain: "No biometric is foolproof."

Jain estimates that as many as 4% of people can´t be covered by fingerprint-ID biometrics systems because of the comparatively blurry quality of their prints. To get around this problem, the U.S. Immigration & Naturalization Service, among others, has begun recording more than one biometric measure -- fingerprints, plus hand geometry or facial recognition.

Jain and his students, Yatin Kulkarni, and Salil Prabhakar, have been funded by IBM for the past three years to conduct a variety of studies related to fingerprint matching, face recognition, and speaker verification.

Jain and his team of researchers have developed an image-enhancement program, which enables them to remove any unnecessary "noise" from a poor-quality thumbprint while extracting its distinguishing features. Finally, the researchers have developed programs that integrate fingerprint images with facial features and speech data.

By doing so, they are providing the computer with additional useful information as it runs through its automated process of elimination, enabling it to perform with increased efficiency as well as accuracy. "We have combined face and fingerprints, and that integration has resulted in better performance than just using face or just using fingerprints," Jain reports.

Although he acknowledges that a number of concerns have been voiced related to the issue of privacy, he asserts that biometrics can actually help protect an Individual’s privacy when properly used.

It is to Professor Jain’s credit that a number of his former students too are making strides in the field of biometrics. Two examples are Nalini Ratha and Sharat Pankaranti at IBM.

Ramesh Chellappa

Ramesh Chellappa heads the Center for Automation Research at University of Maryland at College Park. He has proposed an approach to recognize humans when they are engaged in some activity such as walking or carrying a package while walking.

With his assistants Naresh Cuntoor and Amit Kale, he is investigating a combination of dynamic features such as the swing of the hands/legs, the sway of the upper body and static features like height, in both frontal and side views. As Chellappa explains: “We know from our experience that gait and posture provide us with cues to recognize people. Consider a familiar person walking at a sufficiently large distance so that the face is not clearly visible to the naked eye. To recognize the person, we may try to combine information such as posture, arm/leg swing, hip/upper body sway or some unique characteristic of that person.”

Medical studies too have shown that that the way a person walks, or gait, is indeed a unique signature of humans and it can be captured by cameras placed at a distance.

The system being developed by Chellappa uses computers and video cameras to identify and track suspects in public places such as airports. The goal is to focus on body dynamics, such as gait, to spot suspects at a distance and to use facial recognition technology to identify them as they move closer to cameras. Its called Human ID at a Distance.

Sargur N. Srihari and Venu Govindaraju

At the State University of New York at Buffalo, computer-science professor Sargur N. Srihari, has turned to computers to verify that each person´s handwriting is unique. Srihari provided the first peer-reviewed scientific validation that each person´s handwriting is individual. He directs the university´s Center of Excellence for Document Analysis and Recognition (CEDAR)

Srihari used artificial intelligence to help develop software that, he says, determines with 96 percent accuracy whether the same person wrote two different handwriting samples. The software applies pattern-recognition techniques to analyze 11 features that make up the overall structure of the writing, such as the layout of the document and spacing of each line, as well as 512 features of individual characters, such as individual stroke marks and the openings and closures of each character.

"We´re trying to push the 96 percent to a higher accuracy," says Srihari.

With $1.3 million in initial funding from a variety of government and private sources, scientists at SUNY Buffalo’s Center for Unified Biometrics and Sensors (CUBS) are designing, developing and prototyping biometric devices for commercialization.

The goal of CUBS is to research and develop customized biometric systems for specific applications, such as homeland security and public health, based on nontraditional biometrics, such as chemical and biological markers, as well as traditional ones, like the shape of the iris and hand geometry.

Such a customized approach differs from the one now dominant in the field, where a single biometric technology is marketed for a range of different applications, explained CUBS director Venu Govindaraju, professor of computer science and engineering.

"For many applications, the more customized a biometrics system is, the more powerful and accurate it will be," he said. According to CUBS scientists, the application should dictate the kinds of sensors needed, how they should be packaged, the level of "intelligence" they require and how much security is needed to transmit the information.

"We believe success in this area depends on being able to combine and ´tune´ technologies to different applications by using contextual knowledge about how the data will be used," said Govindaraju. "The technology for these applications exists. Now it´s a question of figuring out how to build the best devices."

Scientists at CUBS say they are developing new methods for the acquisition of biometric data, as well as its processing and interpretation—efforts that usually are not explored together.

CUBS already is pioneering the exploration of new biometrics, such as chemical and biological sensors targeted to gather data on individuals by detecting and quantifying the presence of various pharmaceuticals and their metabolites, toxins, blood type and even chemical residues on the skin.

Other Indian Americans involved in Biometrics include Dr. Rama Krishnan, a Senior System Analyst at Lockheed Martin Information Technology group working on the Department of Homeland Security (DHS) Biometrics Program. He has served as the technical lead for several biometrics projects including benchmarking of fingerprint matchers, evaluation of new technology fingerprint scanners, and research studies on biometric technology. Dr. Krishnan has worked in the area of biometrics technology and identification applications for over 25 years.

Dr. Ganesh Ramaswamy joined the IBM Thomas J. Watson Research Center in 1995, where he is now the Manager of the Conversational Biometrics Group in the Human Language Technologies Department. His research interests include pattern recognition, statistical signal processing, machine learning, system identification, user modeling and personalization, voice biometrics, multi –biometrics fusion, speech coding, data compression, wireless communication, speech recognition, language modeling, natural language understanding and dialog management.

Meanwhile the rich business of protecting America against terrorists is about to get a lot richer, observed Forbes magazine in an article Feb 20. Three consortiums are bidding for a U.S. government contract that could be worth up to $20 billion over five to ten years, the magazine said.

Legal expert Raj Nanavati and his brother Samir Nanavati, founding partners at IBG, is one of the biometric industry´s foremost experts in the use of biometrics in public and private sector applications. They say that despite biometrics´ many weaknesses, sales could grow quickly, rising more than 500% from 2002 to 2007, reaching revenues of $4 billion. The biggest chunk will come from U.S. government programs that have finally made it through the homeland security pipeline.