Home > News > Industry News > Anatomy: Face advantages and c.....
Browse Categories
New Arrival(26)
Hot Products(10)
Reader & Modules Series(18)
Handheld Terminal Series(14)
POS Terminal(5)
Time Attendance Series(12)
Access Control Series(5)
Sales Promotion(35)
Door Lock Series(11)
Camera Series(11)
Explosion Models
Contact us
Tel:+86-23-67305242 :
Mobile:+86-13667681778
Email:info@hfcctv.com
Web:http://www.hfteco.com/
Facebook:Huifan Technology
Contact Now

News

Anatomy: Face advantages and challenges

    Face recognition technologyDevelopment in China started in the late 1990s, experienced a "technical introduction - professional market introduction - Technical Perfection - applications - all industries use the" five stages. So far, with the domestic face recognition technology continues to mature, the technology is increasingly being extended to the field of security, extendingAttendance,Access controlAnd other products, product line of more than 20 kinds of types, can fully cover the coal mines, buildings, banks, military, social welfare and security, e-commerce and security and defense and other fields.
   Face recognition Advantages
   Compared to other identification technologies, face recognition is more obvious advantages, mainly in three aspects:
   First, the nature of the so-called naturally refers to when the same biometric identification mode with humans (and other organisms) utilized individual identification, by observing and comparing the face distinguish confirm the identity; identification with naturalness as well speech recognition and shape recognition, andFingerprint recognitionandIris Recognition, Etc. due to humans or other organisms can not pass such biometric identification of individuals so I do not have a natural resistance.
   Second non-mandatory, is recognized face image information can actively get tested undetected individual, face recognition using visible light to obtain a face image information, which is different from fingerprinting or iris recognition need to use an electronic pressure sensor to collect fingerprints or by using infrared capture iris images, these special acquisition mode can easily be detected, and thus can be camouflaged with deceptive.
   Three non-contact, compared with other biometric technologies, the recognition of non-contact, direct contact with users and devices do not need, while at the same time be able to meet multiple human faces in the practical application scenarios sorting, judgment and recognition.
   Face recognition technologyApplications are very broad, currently in the country, government and financial industry is a major customer recognition of the market, these institutions have high security requirements, and strong demand for these two industries also contributed to the face recognition market The rapid development. Face recognition market, in addition to the security industry demanding financial sector, public security prisons, the government sector, border control, etc., some ordinary users have gradually begun to accept and use of related products in the emerging information security applications, face recognition technology can provide a more secure and reliable means of identification to use, so as to enhance the safety performance of the entire network information system to effectively curb the criminal activities of various types of networks. Currently, the face recognition technology can be easy with the original technology in the traditional security field to achieve tighter integration, greatly enhance the intelligence, security and ease of use of the original system, expanding the application fields of the original system, thus promote the technological upgrading of traditional industries. From the market perspective, the development of space recognition technology is not to be underestimated.
   Market size of face recognition systems largely related with population density, while China is a high population density countries. Although the domestic face recognition technology and product application technology level there are already relatively well, but the market acceptance is still in early stages of development and other issues, but also requires a lot of market development to enhance market awareness. Compared to a few years ago before but in recognition of the research and development of security companies in an awkward position has been significantly improved. A few years ago, many enterprises have developed the chip and algorithms, but these technologies are not recognized by the market, companies rely on technology to survive, there is the view that dueBiometricsMarket immaturity and lack of market demand, leading companies can no longer focus on technology; also with a view, sufficient market demand, but the level of face recognition technology is still unable to meet market needs, which leads to these companies in trouble.
   Starting in 2008, the market scale recognition only gradually revealed, the pace of development in recent three or four of them are also very fast. A few years ago, government attention, and its importance in this regard is not high, in recent years, government efforts to promote and attention has been strengthened, the development of things, safe city construction development are good examples, but any development are the market needs to undergo a process of knowledge to market acceptance.
Face challenges in the application
   From a practical test run, the gap between user expectations between the current level of technology is still relatively large. Face recognition technology in dynamic pressure monitoring applications actually facing relatively large.
   1. The user wishes to correct alarm rates demanding. The reality is that in theory must accept the high false alarm rate. On the technical side, to achieve high correct alarm rates can be achieved by lowering the threshold, but the cost of lowering the threshold are: high false alarm rate. In order to achieve 95% correct alarm rates, many algorithms may produce 300% or higher rate of false positives.
   2. The user wishes to monitor the database is large enough, often require tens or hundreds of thousands, even millions watch list, hoping to catch the "big fish." The reality is that a large amount of storage capacity would have to accept the high false alarm rate.
   3. Users want a large-scale network construction, it is possible to outline the activities of the trajectory monitoring personnel. The reality is that the high investment necessary to re-build private networks, and associated hardware.
   The user wishes to make use of existing monitoring equipment (cameras, and network). The reality is that the current lack of clarity of the camera, image quality is poor, the video for scene monitoring human face is too small, resulting in network bandwidth is not enough, and so can not use the existing equipment.
   5. The user does not want to produce less false positives even produce false positives. The reality is that so will lose the right to monitor alarm rate and reduce storage capacity, and contrary to the idea of ​​the user.
   6. Light Problems
   Facing various tests ambient light may appear side light, top light, backlighting and high light and other phenomena, and there may be different light each time, even in the light of each position within the monitored area are different.
   7. Face Pose Problems and accessories
   Because the non-fit monitoring, monitoring personnel monitoring area through a natural posture through, and therefore may appear profile, bow, it looked like the face of a variety of non-positive attitude and wearing a hat, black-rimmed glasses, masks and other accessories phenomena.
   8. The camera's image problem
   Many technical parameters affecting the quality of the camera video image, these factors are photoreceptors (CCD, CMOS), the size of the photoreceptor, the processing speed of the DSP, the built-in image processing chip and lens, etc., while some of the built-in camera setting parameters also affect the video quality, such as exposure time, aperture, dynamic balance and other parameters.
   9. dropped frames and humiliating problem
Identification required network computing system to identify and may cause the video frame loss and shame phenomenon, especially in the area to monitor the flow of people, because of bandwidth issues and network transmission capability problems often cause frame loss and shame.