With the acceleration of the construction process of safe cities and smart cities, a variety of security equipment and technical means can be seen everywhere. In particular, face recognition technology has developed rapidly. From traditional physical security to network security to APP and other terminal applications, the use of less than Ten years. Compared with past industry applications, intelligent analysis and recognition gradually progresses from the background alarm to the front desk certification development, and uses more direct, more exposed, and simpler results to display an early warning. The birth of this application has fundamentally solved the fraud, impersonation, The occurrence of crimes such as counterfeiting, from "control" to "defense", is the most fundamental change in the current social application.
First, the application of innovation performance In recent years, the biggest innovation is face recognition "formula" to "formula" from the development. With the breakthrough of technology (light, angle, attitude, etc.), face recognition presents an industry division. Through self-learning intelligent algorithms, the system automatically calculates and “learns” more suitable for an environment, a certain change, a certain characteristic. The "like brain algorithm", even if the algorithm inventor, can not know the specific reason of the algorithm. Therefore, with the development of deep learning technology, face recognition technology has developed on a quantum level. The transformation from product intelligence to system intelligence brings about the intelligence of big data, thereby realizing data analysis and construction of smart cities. Smarter, more accurate, and more convenient service for the common people. From this point of view, the future face recognition product form should tend to "five", namely, miniaturization, simplicity, terminal, network, and intelligence. "Miniaturization" refers to products that are small enough to be integrated; "Simple" refers to simple operations that can be accomplished with a single key or no keys; "terminal" means that you can carry it with you; "networking" means that you can do it at any time without boundaries. "Intelligent" means that the system is universal and can instantly respond to instructions intelligently, such as smart wearable devices.
Second, the problem of development
In the age of mobile Internet intelligence, face recognition detection technology has been very mature and used in common electronic digital devices, detection technology has entered the industrialization of development, but face recognition technology due to price, market, technology, hardware, network Such restrictions are still in the process of industrialization. Face recognition technology is the most advanced application of face recognition. Due to its great technical difficulty and strong pertinence, it is still mainly used in professional fields such as airport deployment and comparison.
1. Impact of face recognition development Under the current trends of cloud computing, big data, and the Internet, the influence and contradictions faced by face recognition development are concentrated on data security, data control, and data standards.
First, the process of learning can be accelerated through cloud computing, but data security cannot be guaranteed by the "cloud."
Second, the emergence of big data provides opportunities for face recognition. At the same time, it is also plagued by "spam". A large amount of useless information reduces the system's judgment process, affects the system's recognition accuracy, and undermines the system's self-learning ability.
Thirdly, there is no unified national standard for face recognition, multiple data types cannot be used, multiple data results cannot be exchanged, and multiple data sets cannot be analyzed. This also affects the integration analysis of the big platform, resulting in a slow success of information construction. .
2. Technical Difficulties In the application of public safety, the current face recognition technology needs to break through three difficulties:
First, the most needed breakthrough is the support of national policies and the formulation of national standards. National policies in some respects represent the latitude of positions and funds. If there is a demand in the industry, if the policy can provide support and give some encouragement in funding, then the market is huge. Standards are even more important in the face recognition industry. For example, if face recognition technology is to be used on mobile phones, it is necessary to unify the camera's photo accuracy, size, size, etc., and the collection source must be unified. In addition, the source must have a unified standard. Second-generation licenses are relatively standard (but not open), and other photo libraries from the Internet or video cannot have good results.
Second, face recognition needs to break through its own technical problems, such as light, angle, and posture. It is still a long way to go from a technical point of view. Therefore, peripheral hardware support is required. For example, fill lights, infrared lamps, camera mounting heights, and matching persons can effectively solve the above problems.
Third, on large database face systems, the accuracy and speed of comparisons need to be further improved. In the era of modern databases, the face database is constantly increasing and increasing. Face technology can be truly applicable, easy to use, and easy to use. It also requires the development of networks, large-scale servers, and terminal data collection quality. Technology itself must rely on the development of society's overall science and technology, and it is currently the most biotechnology that has been put into mass production in society.
Third, application potential and trends
With the popularization and replacement of second- and third-generation ID cards, coupled with the “one-to-one authentication” required by various industries, it has brought tremendous business opportunities to face recognition, including *, national security, telecommunications, customs, transportation, and finance. There is a wide range of demands in industries and sectors such as social security, social security, medical care, education, industry and commerce, industry credit, and other civil security controls. It has extremely broad market application prospects.
At present, the reconnaissance methods of the Chinese government are mainly criminal investigation, technical investigation, network investigation, and map detection. With the increase of Pingan City's monitoring network, the map detection method has become the main reconnaissance method. Inquiries about the identity of criminal suspects with document data that are counterfeit or do not have identity documents are widely used in criminal investigation, security, patrol officers, and household registration. Using surface image recognition technology, only the face image of the suspect is photographed, and the facial image data is transmitted to the computer center database through the computer network. The system can then automatically compare the fugitive faces in the face database to make an identity judgment quickly and accurately. In recent years, the "Smart City" has also boosted the application of face recognition.
With the development of China’s social economy, various types of economic activities have become more frequent, and the number of migrants has increased rapidly. Various types of counterfeit identity documents have flooded and flooded. With the help of face technology, the security and correspondence of identity documents can be fundamentally achieved, and face recognition can be achieved. It also provides powerful technical support for the second-generation ID cards that have been promoted by the current household registration management department. Calculated by the total population of 1.376 billion in mainland China, this market capacity is undoubtedly enormous.
There are approximately * bureaus in the country, more than 80,000 police stations, nearly 9,000 prisons and detention centers, and 220,000 financial institution outlets. The number of Internet users has reached 668 million, the number of Chinese mobile phone users is 1.306 billion, and there are 291 prefecture-level cities in the country, with 358 counties. County-level cities initially estimated that in the related industries in China (such as *, Guo'an, *, finance, etc.), the total market capacity of face recognition series products is more than 10 million sets. If you add business and civil, the market Worth trillions. This does not include overseas markets and service fees. Therefore, it should be clearly understood that face recognition is a technology with great growth potential, and that such potential applications are still in early childhood, so every effort made by the security industry will be exceeded in the future. Return.
In computer information system security control, e-commerce, e-government and many other aspects, there is an urgent need for identity authentication technology with excellent anti-counterfeiting and anti-counterfeiting capabilities to replace the currently widely adopted user account/password/digital certificate model. According to the National Institute of Standards and Technology/Federal Information Processing Standard, “Although relying on passwords is a traditional method of identity authentication, the security of this method is not satisfactory. In this case, it needs to be more advanced and safer. A higher-غير مجاز مي باشدt, more غير مجاز مي باشدt-effective authentication method to meet the needs of all aspects of society.Biometric identification technology, especially good-practice face recognition technology, undoubtedly provides a practical and comprehensive solution for a large number of users. The generation and replacement cycles of three generations of certificates will be used to predict that biometric technology will become the most important technology in the IT industry in the next 30 years.
magazine book brochure
Advertising Banners,Business Cards Co., Ltd. , http://www.chsticker.com
برچسب:
،
ادامه مطلب
بازدید:
