Facial Biometrics: Identifying & Verifying Individuals from Digital Images

Facial biometrics refers to a technology system that has the capability of identifying as well as verifying a given person form a digital image that is stored in a database. The system works through a comparison of select facial features from an image being verified with the faces from a connected database (Kumar & Manro, 2018). Therefore, it can uniquely identify an individual by analyzing the patterns existing in personal facial shape and textures. There are various goals and impacts of technology. They are based on mobile platforms and deployment in security services.

Is your time best spent reading someone else’s essay? Get a 100% original essay FROM A CERTIFIED WRITER!

Save 25%

Concerning the goals of the application of the technology in mobile platforms, it is applicably used in social media, ID verification, as well as Face ID (Kumar & Manro, 2018). Starting with social media, there has been the adoption of the facial biometrics to aid in the diversification of functionalities to attract a widened user base during stiff competition from different applications (Peng, Spreeuwers, & Veldhuis, 2019). For the case of ID verification, the technology is used in banks and e-businesses to ratify that indeed, a given ID belongs to the holder at the time, for security purposes (Kumar & Manro, 2018). Finally, for the use of the technology in face ID, it is used by such tech companies as Apple to verify the ownership of various phones by the users who claim ownership, thus guarding against theft. There is the use of facial structure, texture, and shape by the biometric authentication, and using facial recognition sensors, a face is verified. Thus, the ultimate goal is to ensure the security of mobile devices handled in the long run.

Besides, there is the goal of the technology to be applied in the deployment of security devices. Commonwealth, United States of America, The Netherlands, China, and South Africa have applied it (Kumar & Manro, 2018). Using the story of The Netherlands as an example of the named countries, it has used artificial intelligence technology and facial Recognition since 2016. The database owned by Dutch police presently contains 2.2 million pictures belonging to 1.3 million Dutch citizens, accounting for 89% of the entire population (Kumar & Manro, 2018). Many thousands of Cameras have been deployed in Amsterdam (Peng, Spreeuwers, & Veldhuis, 2019). In such a case, there is the recording and keeping of faces in database systems, after which the people are verified using facial specimens in the database once in such areas as banks in which security of money is of greatest concern(Kumar & Manro, 2018). Therefore, it can be concluded that facial biometrics technology is used to ensure security in such areas as banks and protection of mobile devices form theft.

The implications of facial biometrics technology can be discussed using two categories of ethics. They include virtue ethics and consequential ethics. Starting with virtue ethics, it is concerned with the setting of some virtues inclined at informing ethical decisions to live a valuable life (Peng, Spreeuwers, & Veldhuis, 2019).. Thus, the application of facial biometrics in diverse social, economic, and political fronts, aimed at promoting security, also contributes to the virtue ethics of individuals by staying away from any social vices lest they are recognized (Peng, Spreeuwers, & Veldhuis, 2019). Finally, for consequential ethics, it is concerned with the net effects of one's deeds. Since it is a security device, people will only be directed to doing good after avoiding vices, where they can then remain safe once in the public domains where the surveillance facial biometrics are in operation.

Vocabulary

Ninhydrin: a chemical for detecting ammonia or primary and secondary amines.

Iodine: the heaviest of halogens, and exists as lustrous nonmetallic solid at standard conditions.

Latent Fingerprints: fingerprints lifted from a surface

Exemplar: fingerprints collected from subjects for enrollment to programs or investigation of a particular crime

Biometrics: metrics related to human characteristics, used for measurement and calculation.

Technology Description

To achieve security, there is the use of three technological solutions. They include the use of facial biometrics, fingerprints biometric technology, as well as the eye iris recognition system.

Facial Biometrics

There are various technological techniques used in the undertaking of face acquisition using technology. In all the techniques, face recognition gets performed in two steps. They include feature extraction and selection, followed by a classification of objects (Soldera et al., 2017). The techniques involved in the two steps are:

3-Dimensional Recognition

The technique makes use of 3D sensors in capturing information concerning the shape and structure of faces. The data is then used in the identification of distinct features on the facial surface, including the chin, nose, and contours of the sockets of eyes. One significant advantage of applying 3D facial Recognition is that it cannot be affected by lighting changes as compared to other techniques (Soldera et al., 2017). Further, it can identify a face from ranges of viewing angles such as a profile view. There are three-dimensional points from the surface of the face that greatly improves the accuracy of Facial Recognition. 3 Dimensional research gets enhanced by the developed series of sophisticated sensors that perform a better computation of 3-D facial imagery (Soldera et al., 2017). Sensors in the technique work through the projection of light to the face. More than a dozen such image sensors can get placed on a similar CMOS chip - with each sensor capturing diverse parts of the spectrum.

A perfect 3 D matching technique can be sensitive to expressions. Thus, there is the use of a group of Technion applied tools rooting from metric geometry in treating expressions, including isometries. There is also a new method used to introduce a manner of capturing 3D pictures by the use of three tracking cameras pointing different angles (Soldera et al., 2017). One of the points at the front part of the subject, the second one pointing to the side, and the third one pointing at an angle. All of them will then work together in tracking the face of the subject quite readily while effectively detecting and recognizing it.

Skin Texture Analysis

Another emerging technique uses visual details of subjects' skins. The skins are captured in scanned or standard images. The technique is also referred to as skin texture analysis and turns some unique patterns, lines, and spots in the skin of the subject into mathematical spaces. It works as atypical Facial Recognition. Pictures are taken and the patches of oasda work by distinguishing the lines, skin texture, and pores in the skin. It has the ability to identify and distinguish the contrast existing between any two similar pairs (Soldera et al., 2017). Through the addition of technology to facial Recognition, the Recognition of faces rise from 20% to 25% (Soldera et al., 2017).

Traditional Technique

In the method, some facial recognition algorithms identify facial features through the extraction of features and landmarks from the face. An example is an algorithm used in the identification and analysis of relative size, position, jaw, nose, cheekbones, and the shapes of eyes. They are then utilized in search of other images having some matching characteristics(Soldera et al., 2017). Other algorithms normalize the gallery of facial images then compresses the data, where the image data useful in facial Recognition is saved. The probe images are then compared with face data in the database.

Recognition algorithms can be subdivided into two categories. One of them is geometric, and it looks for features that distinguish the face. The second one is photometric, which acts as a statistical approach that works by distilling a given image to values then compares them with templates in the run to eliminate variances (Soldera et al., 2017). The algorithms in such a process can be classified into two broad categories. They are feature-based and holistic models. Feature-based models subdivide the face into components, including according to features followed by analyzing them alongside their spatial locations concerning other features recognized by the technique (Soldera et al., 2017). Some of the most used recognition algorithms are principal component analysis that uses eigenfaces, elastic bunch graph matching, and linear discriminant analysis that uses the Fisherface algorithm, as well as the multilinear subspace learning and hidden Markov model that uses tensor representation.

Thermal Cameras

It has been made through the amalgamation of Skin Textual Analysis, 3D Recognition, and traditional technique to result in it having a higher success rate. Since it is a combined technique, it has one merit over other techniques. It is comparatively insensitive to the changes in the expressions, including smiling, frowning, and blinking (Soldera et al., 2017). It also can compensate for beard growth or mustache alongside the appearance of eyeglasses. It is also uniform in respect to gender and race.

Thermal cameras, while in use, will mainly detect the head's shape and ignores subject accessories, including makeup, hats, and glasses. They can capture facial imagery even under low lights or nighttime conditions without the need to use the flash that exposes that position of the camera. However, it has a disadvantage where the database for facial Recognition is quite limited. The technique works via the combination of local information - features about eyes, mouth, and nose and the global information - features regarding the entire face (Soldera et al., 2017). Alongside its enhancement of the discriminatory nature of the analyzed image, the technique can get used in the transformation of thermal face signature to the refined and visible image of the face being handled.

Fingerprint Biometric

The technological system has the same purpose of use in security as facial biometrics. The technology detects the impressions that have been left by friction ridges of human fingers. Their recovery in a crime scene is crucial for security activities of tracing down the likely suspects (Lee et al., 2017). It works by the principle that human fingerprints are detailed, difficult to change, nearly unique, as well as durable, hence making them suitable for use as long term references of human identity they can be utilized by authorities such as police in the identification of subjects that wish to hide their identities. They can also be employed in the identification of the deceased or the incapacitated who cannot identify themselves, such as during the aftermath of a given natural calamity such as an earthquake (Lee et al., 2017). The following are the two fingerprinting techniques:

Latent

A latent fingerprint refers to a partial fingerprint that was lifted from a given surface. Grease and moisture on human fingers lead to latent fingerprints on such surfaces as glass. Since they are not quite visible, they can be detected using chemical developments of powder dusting, soaking in silver nitrate, iodine fuming or spraying of ninhydrin(Lee et al., 2017). Different methods of chemical developments are used depending on the surfaces in which the fingerprints have been left. For porous surfaces, there is the use of paper, and for non-porous surfaces, there is the use of plastic, metal, or glass (Lee et al., 2017). Dusting processes are required for non-porous surfaces, in which fine powder alongside brush is used, together with the application of some transparent tape for lifting latent fingerprints off the surface.

Exemplar Technique

It is applied for the fingerprints that have been collected deliberately...