In the past few weeks, three global high-tech companies (IBM, Microsoft, Amazon) have announced that they will not support the use of new facial recognition technology by law enforcement agencies such as the police. In fact, it wasn’t the other that announced the news at IBM, it was by its chief executive, which highlighted the potential damage to the technology in terms of racial identity and human rights violations.

Recently, the United States introduced a bill in the Senate facial recognition and biometric technology suspension act to prohibit law enforcement agencies from using facial recognition technology.

Some people believe that face recognition technology has racial bias. In the analysis and identification of personal goals, technology can provide very accurate results for white individuals (male / female), but for black people (mainly women), the effect is very low, so there are problems of technology bias and related challenges. This may be due to the low availability of data sets and training models. Once the technology obtains a large amount of data and uses the correct algorithm, it can solve the problem.

The global face recognition market is expected to grow by 14.5% CAGR in the next seven years

Face recognition has three main components: computer vision, AI based algorithm and database for training and verification. The camera captures the image and maps it to the data repository to identify the target. It mainly works on the markup function set, which establishes correlation, data accumulation and aggregation. Parameters, such as the distance between eyes, nose and face shapes, define more than 100 data points, which are defined on the data matrix.

Facial recognition is an important part of our daily life, which is a common unlocking function of our mobile phones. We can see it in the airport for security authentication and access control in high security places. Although there are other biometric scanning technologies, such as iris scanners and fingerprint scanners, these technologies are more useful for small-scale personal identification. Face recognition is used in a wide range of people recognition.

In law enforcement agencies, face recognition is widely used, mainly for large-scale surveillance, crowd management, personnel counting, etc., which is more common in smart city management.

According to the research of global market research company grand view research, the global face recognition market is worth 3.4 billion US dollars and is expected to grow by 14.5% CAGR in the next seven years. Based on technology, the face recognition market is divided into 2D, 3D and facial analysis.

In Israel, the United Kingdom, Germany and other countries / regions, face recognition technology has been widely used for large-scale personnel detection, especially in the current epidemic of new coronavirus, many regions have launched larger scale monitoring programs, such as deploying more than a large number of built-in face recognition technology and other video analysis applications in public areas of major cities.

For the collected and stored data, there are also security and privacy issues. The events of data leakage, cloning and adjustment have been fully verified by facts. Currently, there are few regulations configured to meet these challenges. Gdpr (general data protection regulation) is one of the regulations formulated by EU to solve the problem of privacy.

In addition to global high-tech companies, many camera OEMs offer fr analysis as part of the bundled solution. Significant share of the global OEM market comes from Asia. In addition, many professional video analysis companies provide face recognition technology as an independent solution.

In India, the share of technology in effective regulation is increasing. The police modernization program was initially launched through the launch of the crime and crime tracking network system (cctns). This is further reinforced by “digital India initiatives” (such as the “safe and smart city” initiative).

According to the government of India’s police research and development agency, as of January 1, 2017, the police to population ratio was 193, compared with the United Nations recommended ratio of 220 (i.e., the number of police officers in 100000 population). This proportion has increased significantly from 142 in 2006 to 193 in early 2017. Technology can further bridge this gap and help police maintain law and order, prevent and mitigate crime.

In major cities such as Mumbai, Delhi and Hyderabad, recent safe city plans include video analysis and facial recognition. Face recognition analysis and prediction is an important element in deploying solutions in these cities. This has produced good results and helped the city quickly resolve major crimes. Police leadership has embraced this technology, which can help them allocate and deploy manpower correctly and monitor effectively.

Editor in charge: Tzh

Leave a Reply

Your email address will not be published. Required fields are marked *