At present, with the rapid development of information technology, biometric technology has been widely used in police combat and personal property protection. Recently, the Chinese Academy of Social Sciences pointed out how to apply face recognition technology to public security and prevent possible legal risks.
Face recognition, face recognition technology, police technology, artificial intelligence
The report of the 19th National Congress of the Communist Party of China proposed to promote the deep integration of the Internet, big data, artificial intelligence and the real economy. With the rapid development of information technology, biometric technology has been widely used in police combat and personal property protection. As a kind of biometric technology, face recognition technology uses its unique advantages to make use of the differences in individual facial features, and is widely used in government, military, banking, e-commerce, security and defense. Recently, the Chinese Academy of Social Sciences pointed out how to apply face recognition technology to public security and prevent possible legal risks.
The transformation of weak artificial intelligence to strong artificial intelligence
Face Recognition (Face Identification Technology) is based on the person's facial features, capture an image or video stream containing the face by a camera or a camera, and automatically detecting and tracking human faces in the image, and thus the detected human face A series of related techniques for face recognition, also known as face recognition, portrait recognition. Face recognition technology usually includes three parts: face detection, face tracking, and face matching. The research of this technology began in the 1960s, with the development of computer technology and optical imaging technology improved after the 1980s, and the real entry into the primary application phase in the late 1990s, and the realization of technology in the United States, Germany and Japan. Mainly. "Face recognition system" integrates various professional techniques such as artificial intelligence, machine recognition, machine learning, model theory, expert system, video image processing, etc., and needs to combine the theory and implementation of intermediate value processing, which is the latest application of biometric identification. The realization of its core technology shows the transformation of weak artificial intelligence to strong artificial intelligence.
Generalized face recognition includes a series of related techniques for constructing a face recognition system, such as face image acquisition, face location, face recognition preprocessing, identity confirmation, and identity lookup. Narrowly defined face recognition refers to the technology or system for identity verification or identity search through face. This paper is based on the application research of the latter. Currently, face recognition, fingerprint recognition, iris recognition, gait recognition, voiceprint recognition, and DNA recognition are the six major biometric technologies. Compared with other biometric technologies, face recognition technology has the following characteristics.
One is non-contact. Face recognition technology does not require direct contact with the device to capture facial images, while most other biometric technologies need to be extracted by individual contact with the device. For example, fingerprint recognition technology has high technical requirements for fingerprint extraction. Each time the fingerprint is printed and the focus is different, the acquired fingerprint will be deformed to varying degrees.
The second is unconsciousness. The face recognition technology does not need to be specially matched with the face collection device, and the individual can acquire the face image in an unconscious state, and other biometric technologies almost all require individual cooperation to complete the extraction of the biological feature. For example, DNA recognition technology requires individuals to cooperate with the collection of blood, saliva, etc. Once mismatched, specimens will have resection sites, sample contamination, etc., affecting the accuracy of the results.
The third is practical. The face recognition system mainly recognizes and calculates the hairline, eyebrow, eyebrow, eye shape, nose shape, mouth shape, etc., and cuts the face screenshot or photo to identify the similarity of the sample (face database). Sorted by high to low by percentage, and can realize sorting, judgment and recognition of multiple faces. The same non-contact voiceprint recognition technology is greatly disturbed by the environment, exhibits variability characteristics in multiple environments, and the practical effect is not strong. The fourth is cost-effective. The horizontal comparison of acquisition cost, comparison efficiency and uniqueness of life characteristics, face recognition system cost-effective. Although the iris recognition technology has high precision and uniqueness, it has high acquisition cost and low recognition efficiency, and is not suitable for a wide range of popularization applications.
Technical applications lack legal norms and standards
According to the Biometrics Market and Industry Report 2009-2014 published by the International Biometrics Group (IBG), face recognition accounted for 11.4% of the global biometrics market in 2014. In the field of domestic police work, face recognition technology has developed rapidly, with an increase of 18% in the past five years, which fully demonstrates that the application potential of this technology is huge. Taking the results of the L City Public Security Bureau's face recognition system in actual combat as an example, as of October 2018, the L City Public Security Bureau's portrait recognition system used the static portrait matching function to query 47,621 times, and used the dynamic portrait recognition platform to issue personnel warnings of 12,000. For the rest of the year, 734 cases of various types of cases were cracked through system application, 416 suspects were arrested, and 85 fugitives were captured by the front end of the portrait recognition, which was nearly four times the amount of escape in the same period, effectively improving the efficiency of police work.
There are some difficult problems in the application of face recognition technology.
First, technical reliability and availability are limited. Due to the possibility of external interference due to facial features, the accuracy of the recognition technology has not yet reached 100%, and there are hidden dangers of false positives and false positives. First, due to the influence of illumination, occlusion, posture, image quality and other conditions, the recognition rate of the face recognition system will decrease. According to reports, two professors from the National Institute of Informatics and the College of Engineering in Tokyo have developed glasses that can block facial recognition by emitting near-infrared light, which absorbs light of a specific wavelength through a light-transmitting material, thereby blocking Break the light path of the face recognition system. Second, changes in age lead to changes in facial appearance, while ID cards are 10 years, 15 years, 20 years, etc., during which individual changes will affect the accuracy of recognition. Third, the special facial groups such as twins have similar facial structures, which have a certain impact on the accuracy of face recognition. Fourth, the lawless elements changed their original facial features after facial reshaping, such as bone removal and cushioning of the nose, which led to a decrease in the recognition rate. Fifth, the technology cannot be directly used as evidence, such as fingerprints and DNA. It is usually an auxiliary means of identification. It needs to be further verified through inquiries and evidence collection, which has certain limitations.
Second, there are security risks and risks. Face recognition technology is expanding in the field of application, and it also exposes certain security risks. The face feature has the characteristics of being unique and unchangeable for life. If the criminals use the acquired user identity for illegal purposes, it will cause leakage or even violation of the citizen's personal privacy. For example, illegally obtaining the characteristics of citizens' biological information, wearing high-definition 3D masks, and coping with system instructions to make corresponding actions, this fraudulent means is very threatening.
Third, there is a lack of laws and regulations and industry standards. Although China's "Network Security Law" explicitly includes personal biometric information into the scope of personal information, the use, storage, transportation, and management of information still need to be further refined. China's current "Public Security Fingerprint Identification Application Image Technical Requirements" (GB/T35736-2017), "Public Security Face Recognition Application Image Technical Requirements" (GB/T35678-2017) and other related standards are all specific to the public security organs. The criminal investigation, license management and other public security services and security based on the development, but with the gradual widening of biometrics in various fields, especially in the financial sector, there is still a lack of uniform industry standards system established from the human biometric technology level.
Protect against security risks and achieve greater effectiveness
First, driven by technological development, the accuracy of face recognition is improved. Accelerate the development of biometrics technology and big data, cloud computing, etc., continuously improve the accuracy of biometric technology, reduce the cost of technology use, encourage enterprises to increase investment and technology research and development, and shorten the cycle of biometric technology from research and development to use. Biometrics technology provides technical support for use in various fields. At the same time, further establish and improve the technical standard system, research and develop technical standards such as biometric basic technology, feature extraction, security encryption, clear the application scenarios and security requirements of biometric information, and guide the improvement of the safety level of biometric identification technology in the industry.
Second, the legal norms are used as a guarantee to prevent the security risks of face recognition technology. Starting from industry norms, further clarifying standards, strict market access, organizing authoritative testing, promoting the healthy development of the industry, effectively preventing some manufacturers from exaggerating publicity for commercial interests, making it difficult for users to distinguish between authenticity and falsehood, and avoiding industrial chaos. At the same time, it is necessary to improve the level of enterprise network security protection, strengthen data storage security, and set up corresponding procedures such as early warning and termination. Once problems occur, they can be discovered and resolved quickly. It is possible to prevent the leakage of original information and protect the privacy of citizens. Vulnerability correction procedures, which cover all kinds of cloaking techniques used by criminals to hide their true identity, identify and crack one by one, and minimize security risks. For the production, sale, purchase, use of other people's masks and the use of illegal means such as 3D printing technology to obtain other people's facial recognition features for illegal interests, to prevent and attack from the legal level.
Third, with the direction of police technology, the face recognition technology will be more effective. With the maturity of face recognition technology, objective conditions for wide application can be realized through terminals such as the Internet and mobile phones. Public security organs at all levels must firmly establish the idea of ​​"taking police force and fighting power into science and technology", constantly updating concepts, popularizing relevant scientific and technological knowledge, learning common operational skills, and applying face recognition technology more widely to arresting and escaping personnel. The illegal personnel, the investigation of unknown corpses, the investigation of missing persons, the comparison of video surveillance, the investigation of witnesses, the identification of identity information, the identification of key populations, etc., save public security work costs, and more easily and effectively crack down on crime and serve the masses.
In short, for face recognition technology, we must be full of confidence, use an open and inclusive attitude to look at the areas that are not perfect, and also maintain a rational and prudent attitude to deal with the arrival of the "brush face" era, to achieve "brushing face" "Do not change face." It is recommended to speed up the establishment of a sound face recognition technology using uniform standards, and formulate corresponding laws, regulations and system specifications, so that face recognition technology develops in a high-security, high-portability, non-contact and low-cost direction.
50H Narrow-body Fuel Dispenser
Double Hoses Fuel Dispenser,Single Product Submersible Pump Type Fuel Dispenser ,Single Hose Fuel Dispenser,Double Products Submersible Pump Type Fuel Dispenser
Henan Wenjia Energy Technology Co., Ltd , https://www.hnwjny.com