Face recognition is software that studies, explores, and then confirms the identity of a person in a picture or a video. Face recognition is one of the most powerful inspection tools ever made. While most people interact with the face recognition software daily, during the opening of their phones and laptops. But, how the government interacts with these face recognition software has a greater impact on the lives of citizens.
When we have the software or device we are using, we can easily stop the software or choose not to open it. But, this case is different when we are in public, because public cameras recognize our faces without taking our consent. This is against the privacy policy of users. Because of this reason most of the big companies like Amazon, Microsoft have refused to give their facial recognition software to governments and big companies because they use the software which is against the privacy law of the public. But, as the software improves its quality and old regulations expire, society will be able to answer big questions that where companies and governments can use software and up to which extent
Working Of Facial Recognition Software
Most people have seen people using facial recognition software in movies. But, their depiction of the use of the software is incorrect. The use of software depends on where it is used by the people and upon which algorithm it is built.
There are a large number of technologies and algorithms upon which facial recognition software is built. But, most commonly we can classify them into three different types.
- Detection-Detection is the process in which software finds a face in an image. If you have seen technology in which a box is made around the face then you have seen the working of detection software. The process of detection only involves the finding of the face and not the identity behind it.
- Analysis– Analysis is the process in which technology maps the face, by measuring the distance between the nose and chin and between the two eyes and then converting this information into different points called “faceprint”. Instagram and Snapchat use similar technology in their apps.
- Recognition– During the recognition phase, technology attempts to confirm the identity of the face behind the camera. This process is used in banks, new mobiles, and financial institutions to confirm the identity of the person behind the camera. If you insert different pictures in the software, soon it will recognize and learn the difference between a face and a wall and between different faces.
The accuracy of the software used during the analysis and recognition process depends on how diverse faces are incorporated into the system during the training of artificial intelligence and technology.
For example, if most of the white men’s pictures are incorporated in the system during the training of the technology. Then, the system will find it difficult to recognize black faces and faces of the women. The best facial recognition software has improved this quality to a greater extent, but still, some technologies find it difficult to recognize faces of the white men.
Once, a company has trained its software to identify the faces of different people. Then facial recognition software will easily find different faces and then compare those faces with its database which is the identification process. During the phase of identification, the software finds the face which is present in front of the camera and then matches that face with the databases available from a variety of sources like pictures from different social media accounts. It then ranks results by different accuracy. This system sounds a bit complicated, but with minor technical skills.
History Of Facial Recognition Software
Facial recognition technology can be traced back to the 1960s when Woodrow developed a system to classify photos of faces. A new face can then be compared against the data points available in the databases. According to modern standards, the system was not fast and efficient enough but the idea had merit.
During the 1970s and 1980s, new approaches were developed with catchy lines like “Fisherfaces” and “Eigenface approach” which improved the efficiency of technology to identify faces and features.
When facial recognition software was first used in the public it brought a big controversy. In 2001, law enforcers used facial recognition software on a large scale on crowds at the Super Bowl. Users find it a violation of the 4th amendment. During the year, some police officials were also found using the same software and then matching the data with national databases.
2010 was the modern era of technology, in which computers were trained enough to recognize faces at a standard feature. In 2011, facial recognition software was used to confirm the identity of Usama Bin Laden. In 2014, Facebook introduced its Deep Face photo tagging software.
Facial recognition was first used in personal devices as a feature of security with Windows and Android phones in 2015, and then with the introduction of iPhone X and Face ID in 2017.
Wrapping Up
Concluding the argument, the question here arises that where the society goes from here is a mixture of policy and personal habits of people. Like, any other technology facial recognition is just software.
But as Mutale Nkonde said: “It’s the way the tool impacts our civil and human rights that is my point of intervention because I think that all technology is agnostic”
