Final Research Paper _ Current and Emerging Technology

Running head: BIOMETRICS I

BIOMETRICS II

BIOMETRICS

Student’s Name:

Course Name:

Professor’s Name:

Abstract

Biometrics is the analysis of an individual’s unique and behavioral characteristics to find his or her identity. The most commonly used technologies are face and fingerprint recognition, speech and voice recognition, gait and DNA matching. All these methods involve four steps: sample capture, feature extraction, template comparison, and matching. It is applied in different industries such as homeland security, health, airport, law enforcement, and education. Face recognition is the most adept biometric technology as it is unique, stable, and does not change over some time. However, certain challenges are experienced while using this technology. This paper reviews different literature sources that cover ways of mitigating the risks in using face recognition devices. These risks include False Reject Rates (FRR).

Keywords: Biometrics, Face Recognition, Behavioural Characteristics, False Reject Rates, Physical Identifiers

Abstract

Biometrics is the analysis of an individual’s unique and behavioral characteristics to find his or her identity. The most commonly used technologies are face and fingerprint recognition, speech and voice recognition, gait and DNA matching. All these methods involve four steps: sample capture, feature extraction, template comparison, and matching. It is applied in different industries such as homeland security, health, airport, law enforcement, and education. Face recognition is the most adept biometric technology as it is unique, stable, and does not change over some time. However, certain challenges are experienced while using this technology. This paper reviews different literature sources that cover ways of mitigating the risks in using face recognition devices. These risks include False Reject Rates (FRR).

Keywords: Biometrics, Face Recognition, Behavioural Characteristics, False Reject Rates, Physical Identifiers

Table of Contents

Chapter 1 IV Introduction IV Research Questions V Chapter 2 VI Literature Review VI Chapter 3 VII Research Methodology…………………………………………………………………. VII Chapter 4……………………………………………………….…………………………. VIII Findings, Analysis, Summary………………………………...…………………………VIII Chapter 5……………………………………………………….…………….…………. VIII Conclusion………………………………………………………….……………………IX References………………………………………………………….…………………………X

Top of Form

Bottom of Form

Chapter 1

Introduction

Biometrics is the measurement and analysis of an individual’s distinct physical and behavioral characteristics. It is used to track terrorists hence often used as the basis of improving national security. The two types of biometric identifiers are through analysis of physical and behavioral characteristics. Physical identifiers include DNA Matching, Face Recognition, Eye -Retina Recognition, Eye-Iris Recognition, Fingerprint Recognition, Voice Recognition. Behavioral identifiers include walking gait, typing patterns, and other gestures.

Factors to consider when choosing the right biometric modality include uniqueness, stability, cost-effectiveness, acceptance, collectability measures, and accuracy. Uniqueness is the measure of how unique a characteristic is among individuals. For instance, using height as a biometric identifier gives a higher chance of identifying individuals with the same height. Stability is also a vital factor to consider when choosing biometric technology. The most stable modality is the use of fingerprints since they do not change with time. The less stable modality is face recognition; appearances drastically change over time as people age. When choosing a biometric technology, one ought to settle for the less costly but highly effective and long-lasting. Several factors contribute to the overall cost, such as durability and the underlying technology.

Certain biometric modalities may be associated with stigma. It is, therefore, necessary to evaluate what is accepted by the users and those that may violate a user’s culture or perception of beliefs. Collectability measures are the efficiency of obtaining data for research. For instance, it is super easy to get a person’s fingerprints as compared to obtaining fingerprints. Several Criteria should be followed when assessing the accuracy of the collected data. They include compliance with standards such as error rate, identification rate, false acceptance rate, and false reject rate.

The goal of this research is to find ways of reducing False Rejection Rates. They occur when an individual tests negative to their biometric template. This can frustrate a user who is not recognized by the system. It may also negatively affect service delivery and productivity. My research will impact the end-users and the staff operating the systems. Providing appropriate training to both the staff and the end-users and ensuring all the protocols are followed to the latter are ways of reducing False Rejection Rates.

Research Questions

1. What are the causes of False Rejection Rates? 

2. What are some of the methods of reducing False Rejection Rates in Face Recognition Technology?

Chapter 2

Literature review

In the book Face Recognition Technologies, the authors' goal is to detect and recognize people captured on camera. They intended to highlight the high level of bias of privacy implication of Face Recognition Systems. The authors used a heuristic with two dimensions, consent status and comparison methodologies, to determine a proposed level of Face Recognition Systems privacy and accuracy (Yeung, Guitierrez &Rand,2020). They then used a more in-depth case study to identify indications of bias and privacy concerns. According to them, the causes of false reject rates were using obsolete training data to perpetuate human bias. Also, using cases in which the subject does not consent to human rights and lack of accessible redress when errors occur in image matching. They concluded that there should be a thorough examination of existing systems and a review of accuracy rates to reduce false rejection rates in face recognition.

In her book our biometric future, Kelly identifies face recognition technologies as a failed technocratic approach to governance where new technologies are seen as short-sighted solutions to complex problems (Gates,2011). She bases her research on press releases, policy statements, and PR kits. Her findings were that face recognition technologies are driven by the priorities of corporations and law enforcement agencies. Therefore, the needs of the end-users are not considered. She concludes that contingency and contestability of face recognition technologies reduce false rejection rates.

In the book Protecting Individual Privacy in the struggle against terrorists, congress suggests that all agencies that collect personal data through face recognition should evaluate their programs' effectiveness, lawfulness, and privacy impacts (Academic Press,2008). This comes after there were reported cases of people wrongfully convicted and their personal information made public. The congress examines two specific technologies, data mining and behavioral surveillance. Regarding data mining, they find that they are less helpful in countering terrorisms. Regarding behavioral surveillance, they find no scientific consensus on whether this technique is ready for operational use in countering terrorisms. They concluded that congress ought to re-examine existing privacy policy, and any individual harmed given meaningful redress.

Biometric Recognition Challenges and Opportunities describes the automated recognition of individuals as a way of identifying terrorists. The authors argue that corporate surveillance has instilled fear in people (Pato,2010). They issued questionnaires to 40 % of the population, and 60 % of the respondents filed and returned them. The findings were that the technologies used in face recognition were complexly heightening the chances of false rejection rate. They also found out that the risks associated with face recognition technologies were inevitable. They concluded that the complexities of the systems needed to be dealt with appropriately.

Harry Weschler explores the challenges facing face recognition and how they contribute to a rise in false rejection rates. They draw their findings from data mining, computer vision, statistics, signal and image processing. Their results were that the main contributory factors included cluttered environments, occlusion and disguise, temporal changes, and robust training (Charles,2007). They concluded that addressing face recognition problems was an adept way of mitigating face rejection rates.

Chapter 3

Research Methodology

To accomplish the stated goal, I plan to encourage the staff and policymakers to examine the systems to avert the negative consequences thoroughly. Thoroughly examining the systems eases the complexity of obtaining a patient's image from the available data. This will, in turn, reduce frustrations and log jams improving service delivery. I would also advocate for implementing privacy policies that protect individuals against false rejection rates and infringing privacy rights. Additionally, I would ensure that individuals receive advanced training on the use of face recognition technologies.

Lowering data points raises false acceptance rates and reduces false rejection rates. A false acceptance rate may cause an organization to accept impostors. Therefore, it would rather have higher false rejection than acceptance rates. False rejection rates may delay service delivery and retard productivity. It also causes the organization to incur additional expenditure on revalidating the users. False rejection rate negatively impacts the population as it causes unnecessary frustrations.

By reducing the false reject rate, an organization can save on costs that would otherwise be used in revalidating users. The organization's reputation will improve as there will be reduced conflicts between the organization and the customer. There will be ease in operating the systems; therefore, staff productivity will be heightened. By reducing false rejection rates, false acceptance rates will rise, leading to giving impostors access to the organization.

Chapter 4

Findings

After comparing the literature reviews, my findings were the use of poor training data and errors in matching images were some of the causes of false reject rate (FRR). Kelly argues that priorities of corporations and lawful agencies in face recognition technologies are the cause of face reject rates. Other researchers like Harry Wechsler reported that complex systems caused FRR. The National Academic Press blamed the FRR on the lack of an effective privacy policy.

Analysis

The researchers attempt to find the causes of false reject rates using different methodologies such as questionnaires, press releases, policy statements, heuristic, and behavioral surveillance. After thoroughly examining the causes, they offer solutions that they think would help mitigate the rise of FRR. Despite the difference in methodologies and findings, they all have one goal, to reduce False Reject Rates.

Summary

In summary, reducing false reject rates have both positive and negative impacts on both the organization and the population. Therefore, appropriate measures ought to be taken to avert the negative impacts.

Chapter 5

Conclusion

In conclusion, face recognition remains a challenge in various sectors. It has received great attention over recent years due to its use in improving national security. It is also used in other sectors such as health care and airport security. A hundred percent efficiency in the use of this biometric technology is yet to be achieved. This is due to the challenges they face, such as the false reject rates, which have led to many false convictions and frustrations among users. Researchers invested their time in the field to find ways of reducing the FRR and advocated for policies that protect against infringement of rights.

References

Top of Form

February, Inc. (2008). Protecting individual privacy in the struggle against terrorists: A framework for program assessment. Washington, D.C: National Academies Press.

Bottom of Form

Top of Form

Gates, K., & New York University Press. (2011). Our biometric future: Facial recognition technology and the culture of surveillance. New York: New York University Press.Bottom of FormTop of Form

Pato, J. N., Millett, L. I., National Research Council (U.S.)., & ProQuest (Firm). (2010). Biometric recognition: Challenges and opportunities. Washington, D.C: National Academies Press.

Top of Form

Wechsler, H. (2007). Reliable Face Recognition Methods: System Design, Implementation and Evaluation. Dordrecht: Springer.

Bottom of Form

Bottom of Form

Top of Form

Yeung, D., Balebako, R., Gutierrez, C. I., Chavkowsky, M., & Rand Corporation. (2020). Face recognition technologies: Designing systems that protect the privacy and prevent bias.

Bottom of Form

References

Top of Form

ebrary, Inc. (2008). Protecting individual privacy in the struggle against terrorists: A framework for program assessment. Washington, D.C: National Academies Press.

Bottom of Form

Top of Form

Gates, K., & New York University Press. (2011). Our biometric future: Facial recognition technology and the culture of surveillance. New York: New York University Press.Bottom of Form

Top of Form

Pato, J. N., Millett, L. I., National Research Council (U.S.)., & ProQuest (Firm). (2010). Biometric recognition: Challenges and opportunities. Washington, D.C: National Academies Press.

Top of Form

Wechsler, H. (2007). Reliable Face Recognition Methods: System Design, Implementation and Evaluation. Dordrecht: Springer.

Bottom of Form

Bottom of Form

Top of Form

Yeung, D., Balebako, R., Gutierrez, C. I., Chavkowsky, M., & Rand Corporation. (2020). Face recognition technologies: Designing systems that protect privacy and prevent bias.

Bottom of Form