Facial Recognition Technology

Facial Recognition Technology

Biometrics technology is gaining recognition due to its ability to maintain high-security levels, convenience, and spoof-proof. Facial recognition technology is a technique used to recognize human faces using technology. It incorporates biometrics to record facial features from a video or photograph. It compares the information with what is stored in a database of other faces to find a match (Kim & Sung, 2009). The technology emerged to address the modern needs of identifying and verifying a person’s identity. The technology fulfills the biometric system requirements that try to recognize an individual’s status by using distinctive body features and functionalities that are more familiar with the implementation of visual surveillance.

The main applications of facial recognition technology include identity detection in security systems, automated roll calls, and student attention and emotion monitoring. Therefore, facial recognition technology can be viewed as an extension of the logic of technology-based surveillance trends. This paper will review the science behind facial recognition, its use in criminal investigations, and the limitations in its application.

The Science Behind Facial Recognition and How It Can Be Used in Criminal Investigations

The facial recognition process includes four connected phases. The first phase comprises detecting a face, the second includes normalizing, the third incorporates extraction, and the fourth consists of recognition. The phases follow the same techniques and rely on one another. Accordingly, they can be described as components independent of a typical facial recognition system. Every phase also poses a significant challenge to the successful operation of the facial recognition system. Identifying a face in an image may be a simple task in the detection phase. Still, computer use becomes more challenging because it decides an image’s pixels (Mohammad, 2020). When an image’s context is filled with objects, detecting the image becomes more complex—traditionally, detecting faces focused on facial landmarks such as the exes detecting the skin color in areas with standard or circular features. The normalization phase includes standardizing the image in size, illumination, and pose compared to the images stored in a database. It also includes accurately allocating major facial landmarks (Zaeri & Cherri, 2009). Face landmarks enable the algorithm to normalize an image to detect the image with different variations. The image variations corrections are based on the inferred statistics and the estimates that are not entirely needed to be accurate.

The probe image must be as close as possible to the even face. All systems need facial landmarks as a critical feature, irrespective of the overall recognition method. The identification process fails if the facial landmark is not found (Zaeri & Cherri, 2009). Normalization helps ensure that similarity is achieved in a lower or higher accuracy degree. The extraction phase includes generating a mathematical representation called a reference biometric or biometric template. The biometric is stored in a database and creates a foundation for the recognition phase. The facial recognition algorithms differ in how they transform and translate the image into a simplified numerical representation to identify the image. In the recognition phase, maximum retention of information is needed to change the numerical representation via biometric analysis.

Facial recognition technology can be used in law enforcement and commercial purposes. In law enforcement, the technology is applied in criminal investigations to stop, resolve, and prevent crimes and ensure that crime victims get justice. In most instances, facial recognition is used to find the identity of criminals in fraud crimes, look for terrorists in public places, and find missing persons. Facial recognition in criminal investigations includes a process whereby security experts input a person’s image into the system’s question and remove unwanted elements from the image. The system then classifies the image based on landmarks such as the distance between the eyes, the shape of the nose and chin, and the jawline. The system then searches throughout the database to find the perfect match and display the results. Law enforcement officers add images to their databases every time a person is arrested, thus creating a record for future reference and comparison.

Challenges to Facial Recognition

The effectiveness of facial recognition technology relies on the ability to deal with various challenges that may affect the credibility of the results. One of the limitations of the use of facial recognition is environmental factors. The main environmental factors are lighting and the type of background. In a natural environment, there are significant differences between lighting settings. It is also difficult to detect a face in complex background settings. The proportion of these factors leads to inaccurate results. The variation in the background can be limited if capturing an image occurs in a controlled environment (Olszewska, 2016). However, if the image is captured in the wild, algorithms need to be able to find the face in a complex environment.

Moreover, variation in illumination affects the performance of the face recognition process and the matching process’ performance. For instance, conventional face detectors that rely on face features fail under severe illumination differences such as overexposure and heavy shadows. Illumination differences can also be limited in controlled environments because the angle between the source of light and the face can create shadows that hide face features, diminishing an algorithm’s ability to make a reasonable identification conclusion.

The second limitation is esthetical factors. According to Rofoo (2017), esthetical factors impact the perception of beauty in facial features. They include color and texture. Procedures such as plastic surgery focus on restoring or correcting the functionality or appearance of visible parts of the human body, such as the face. The templates and markers used in the facial recognition process are affected by esthetical factors. Plastic surgery may conceal the features used to identify a face, leading to false results. Blows and bruises can also temporarily alter the structure of the markers in the face, thus affecting the face identification process. In addition, aging also changes the face’s texture, affecting facial recognition, which could lead to false results. The essential elements of the face are the underlying bone structure that shapes the face, soft tissue, and skin.

Consequently, these elements are affected by aging, which arises from various forces such as skeletal modeling, gravity, hormonal imbalance, subcutaneous redistribution of fat, smoking, and too much exposure to sunlight. Age-related features go against permanence, one of the features that physical characteristics must have to be valid as biometric identifiers. The permanence of a biometric identifier means that it does not change over time. Biometric identifiers must also fulfill uniqueness, universality, performance, collectability, circumvention, and acceptability.

Hypothetical Example of Facial Recognition Being Used in a Criminal Investigation.

Last year, my cousin lost his debit card, and on the same day, the card was used to withdraw money from an ATM in a nearby town. My cousin reported the matter to the police, who began tracing the transaction. The police identified the specific ATM where the card was used and reviewed the CCTV footage. Fortunately, the CCTV camera had recorded the perpetrator’s face. The police compared the image from the camera with other images stored in their database and found a match. The perpetrator was in their database and had a criminal record for fraud and armed robbery. He had just been released from prison and was defrauding people to get back on his feet. The police later traced his phone and arrested him.

Conclusion

Advancements in technology have played a significant role in enhancing security. Facial recognition is currently one of the most embraced technologies due to its ability to verify a person’s identity. The technology includes four phases: detection, normalizing, extraction, and recognition. The main applications of the technology are in law enforcement and commercial settings. In law enforcement, the technology is used in criminal investigations to identify criminals and confirm the identity of individuals accused of a crime. However, the application of facial recognition technology is affected by limitations such as aesthetic and environmental factors. Environmental factors include lighting and the image’s background, while esthetical factors include the elements that affect the perception of beauty. These two limitations challenge facial recognition by interfering with the physical features and landmarks to determine whether two faces match. Therefore, there is a need to improve the algorithms used in facial recognition to reduce the impact of environmental and esthetical factors.

References

Kim, D., & Sung, J. (2009). Face recognition. Automated Face Analysis, 163-254. https://doi.org/10.4018/978-1-60566-216-9.ch005

Mohammad, S. M. (2020). Facial recognition technology. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3622882

Olszewska, J. I. (2016). Automated face recognition: Challenges and solutions. Pattern Recognition – Analysis and Applications. https://doi.org/10.5772/66013

Rofoo, F. F. (2017). Enhancing facial features by using clear facial features. AIP Conference Proceedings. https://doi.org/10.1063/1.5004321

Zaeri, N., & Cherri, A. (2009). High performance and fast face recognition technique based on components of phases of face images. SPIE Proceedings. https://doi.org/10.1117/12.818126

ORDER A PLAGIARISM-FREE PAPER HERE

We’ll write everything from scratch

Facial Recognition Technology

Biometrics continues to be an emerging field, and its use continues to evolve in criminal investigations. As you know, fingerprints are scientifically valid evidence that various courts have accepted. However, many of these new technologies have not had enough time to be scientifically recognized as valid means of identifying somebody for criminal trial-related purposes. With this in mind, examine some challenges with the less accurate forms of biometrics, including facial recognition, voice recognition (voice stress analysis), or signature recognition. Prepare a 5-page paper that addresses the following questions:
• Select 1 of the above-listed less accurate forms of biometrics and summarize the science behind it (i.e., how it works) and how it can be used in criminal investigations.
• Identify at least two challenges to the selected biometric. In other words, what are the limitations of its use?
• Provide a hypothetical example of the selected biometric used in a criminal investigation.
• Support your work with properly cited research and examples of the selected biometrics applied in the public and private sectors.