This week’s assignment was to conduct a mini analysis of a surveillance media or technology. I chose facial recognition, a technology capable of matching a human face from a digital image or a video frame against a database of faces, typically employed to authenticate users through ID verifications. It works by pinpointing and measuring facial features from a given image or video frame. Unlike many other biometric systems, facial recognition can be used for general surveillance in combination with public video cameras, and it can be used in a passive way that doesn’t require the knowledge, consent, or participation of the subject.
Development began on similar systems in the 1960s, beginning as a form of computer application. Since their inception, facial recognition systems have seen wider uses in recent times on smartphones and other forms of technology, such as robotics. Because computerized facial recognition involves the measurement of a human’s physiological characteristics, facial recognition systems are categorized as biometrics. Although the accuracy of facial recognition systems as biometric technology is lower than iris recognition and fingerprint recognition, it’s widely used due to its contactless process.
While humans can recognize faces without much effort, facial recognition is a challenging pattern recognition problem in computing. Facial recognition systems attempt to identify a human face, which is three-dimensional and changes in appearance with lighting and facial expression, based on its two-dimensional image. To accomplish this computational task, facial recognition systems perform four steps. First, face detection is used to segment the face from the image background. In the second step, the segmented face image is aligned to account for face pose, image size, and photographic properties such as illumination and grayscale. The purpose of the alignment process is to enable the accurate localization of facial features in the third step, which is facial feature extraction. Features such as eyes, nose, and mouth are pinpointed and measured in the image to represent the face. Finally, the established feature vector of the face is then, in the fourth step, matched against a database of faces.
Facial recognition is more popular than you think. Facetune and Perfect365, apps commonly used by social media users to edit their photos, use facial recognition systems. Snapchat filters use face detection technology to layer a 3D mesh mask over the face. DeepFace, a deep learning facial recognition system created by a research group at Facebook, was trained on four million images uploaded by Facebook users. The system is said to be 97% accurate, compared to 85% for the FBI’s Next Generation Identification System. In February 2021, TikTok admitted guilt to a US lawsuit which alleged that the app had used facial recognition in both user videos and its algorithm to identify age, gender, and ethnicity.
The emerging use of facial recognition is in the use of ID verification services. Many companies and others are working in the market now to provide these services to banks, ICOs, and other e-businesses. For example, my banking app uses ID verification to allow me to log in without using my password. Android, Apple, Microsoft’s Xbox 360, and Windows 10 also allow facial recognition as a means of unlocking devices.
The biggest danger is that this technology will be used for general, suspicionless surveillance systems. Facial recognition systems are already employed throughout the world today by governments and private companies. State motor vehicles agencies possess high-quality photographs of most citizens that are a natural source for face recognition programs and could easily be combined with public surveillance or other cameras in the construction of a comprehensive system of identification and tracking. Facial recognition can be used not just to identify an individual, but also to unearth other personal data associated with an individual, such as other photos featuring the individual, blog posts (hello government), social media profiles, internet behavior, and travel patterns. Concerns have been raised over who would have access to the knowledge of an individual’s location and people with them at any given time. Individuals have limited ability to avoid face recognition tracking unless they hide their faces.
Moreover, facial recognition systems have a tendency to be both sexist and racist. The error rate for gender recognition for women of color ranged from 23.8% to 36%, while for white men it was between 0.0 and 1.6%. Overall accuracy rates for identifying men were higher than for women at 91.9% versus 79.4%, and none of the systems accodomated a non-binary understanding of gender.
Because of the large risk of error and privacy invasion, all of the benefits of facial recognition can be countered. For example, facial recognition is used by law enforcement agencies to identify criminals and to find missing persons. However, what if the system makes a mistake and confuses you with a criminal? What if in the search for a serial killer, the system mistakenly identified you and you were arrested in front of your coworkers or neighbors? Suddenly, you’re the most hated person in America, your reputation is ruined, and you’re in jail for a crime you didn’t commit. Even worse, you justly resist arrest and then get shot and killed. The error might be fixed, but that doesn’t take away the humiliation. Banks and phones use ID verification to prevent fraud, but one doppelganger steals your phone and suddenly they have your entire life savings at their finger tips. All of your hard work is gone and you end up homeless, even if you did everything right. It’s not fair. Not to mention, it’s just creepy.
It’s not even trusted organizations that make money off of stalking you. Third party tech companies sell their facial recognition software to the government and law enforcement agencies. FindFace is a face recognition technology developed by the Russian company NtechLab. SenseTime is a Chinese artificial intelligence company that develops technologies including facial and image recognition. DeepCam is another Chine computer vision company focusing on face recognition. AnyVision, an Israeli startup, is one of the leading visual AI platforms in the world that develops facial recognition software. Your local police might have an obligation to ethics, but do these international tech companies operate by those same morals? Are you comfortable with never being anonymous for the rest of your life?
The American Civil Liberties Union has campaigned across the United States for transparency in surveillance technology. In May 2019, San Francisco became the first major US city to ban the use of facial recognition software for police and other local government agencies’ usage. In June 2019, Somerville, Massachusetts became the first city on the East Coast to ban surveillance software for government use, specifically in police investigations and municipal surveillance. Since then, municipal use has been banned in Berkely and Oakland, Boston, and Portland, Oregon.
Before you decide what side of history you want to be on, be aware of all the facts.
Until next week,