Foshvision Technology Co.,ltd

Foshvision can supply all high quality security night systems and thermal systems!

Home > News > Content
Seven Daan-proof Biometric Identification Technology Analysis
- Nov 27, 2008 -

So-called biometric technology is, by computer and optical, acoustic, biological sensors and bio-statistics and other high-tech means to close, inherent to the use of human physiological characteristics, (such as a fingerprint, face, Iris, etc) and behaviors (such as handwriting, voice, gait, etc) for identification of personal identity.
Biometric technology based on human characteristics such as fingerprints, voice, etc to identify the individual identity. At present, there are a lot of biometrics can be used for authentication. Here, we describe the most popular biometric technology is how it works, and they capture, extracting features, functions and alignment done by a simple comment.
1, iris recognition technology
IRIS is a fabric in the pupil in the eye-shaped colored rings, every IRIS contains a unique Crown, Crystal, fine wire, spot, structures, pits, Ray, wrinkles and stripes characteristic of structures, it was claimed, no two IRIS are the same. Iris scan security system consists of an automatic camera to find IRIS and if your eyes began to focus, trying to cheat the system by winking is not enough.
Advantages: easy to use; may be the most reliable biometric identification technology, although it has not been tested just before a user located on the device without the need for physical contact.
Shortcomings: a most important of shortcomings is it no for had any of test, current of Iris recognition system just with statistics principle for small scale of test, and no for had reality world of only sex certification of test; hard will image gets equipment of size miniaturization; for focused of need and need expensive of camera, a such of camera of minimum quotes for 4000 dollars; lens may will makes image distortion and makes reliability was reduced; black eyes very difficult read; need a compared good of light.
2, retinal recognition techniques
Retina is a biometric characteristics, some people think that the retina is more unique than IRIS biometrics, laser retinal recognition techniques requirements to obtain the characteristics and uniqueness of the retina on the back of the eyeball.
Advantages: retina is a very fixed biometrics because it is "hidden", so can't wear, aging or disease; users do not require direct contact with equipment;
Is one of the most difficult to cheat the system because the retina is not visible and therefore will not be forged.
Disadvantages: technology without the use of any test of the retina. Obviously, the retinal technology users may bring damage to health, which require further study; for consumers, the retinal technologies unattractive; it is difficult to further reduce its costs.
3, facial recognition
Face-recognition technology based on facial features and the relationships between them to identify, identification technology based on these unique characteristics are very complex, which requires knowledge of artificial intelligence and machine learning system, two technologies used to capture facial images for standard video and thermal imaging technology. Standard video through a standard of camera intake facial of image or series image, in facial was capture zhihou, some core points was records, for example, eyes, nose and mouth of location and they Zhijian of relative location was records down then formed template; hot imaging technology through analysis by facial of capillaries of blood produced of hotline to produced facial image, and video camera different, hot imaging technology does not need in better of light conditions Xia, so even in dark situation Xia also can using. An algorithm and a neural network system with a conversion mechanism can be a fingerprint image into a digital signal, final matches or does not match the signal.
Advantages: facial recognition is a non-contact, users do not need direct contacts and equipment; although you can use desktop video camera, but only the more advanced cameras can be effective high-speed capture facial images;
Shortcomings: using who facial of location and around of light environment are may effect system of accuracy; most research biological recognition of people are recognized facial recognition is most not accurate of, is most easy was cheat of; facial recognition technology of improved rely on Yu extraction features and than on technology of improve, and collection image of equipment will than its technology expensive have more; for for human facial of as hair, accessories, variable old and other of changes may need through artificial intelligence to get compensation, Machine learning function must be constantly before the image and now be compared; to improve core data and make up the tiny difference; it is difficult to further reduce its costs, we must be at a high cost to sell high quality equipment.

4, signature verification
Signature as a means of authentication has been used for hundreds of years, and we are all very familiar with the format in the form of a signature in a Bank as a symbol of our identity. Digital signature is a process, and measuring the image itself and the entire signing – in each letter and the order of letters between different speeds, and pressures, signature recognition and voice recognition, is a kind of behavioral science.
Advantages: use signatures to identify more easily accepted by the public and are a recognized form of identification technology.
Shortcomings: with experience of growth, temperament of changes and lifestyle of change, signature also will with and change; to processing signature of not new from of natural change, we must in security aspects do to compromise; because signature of speed unpleasant, we cannot in Internet Shang using it; for signature of handwriting Board structure complex and price expensive, because and notebook computer of touch Board of resolution has is big of differences, we in technology Shang hard will both combined up; hard will it of size miniaturization.
5, voice recognition
Voice recognition technology and the same signature recognition, voice recognition is also a behavioral recognition, voice recognition devices continuously measure, record sound waves and changes. And voice-recognition based on data to the sound voice templates are registered with an exact match.
Pros: voice recognition is also a contactless identification technology, users can quite naturally accepted.
Shortcomings: and other of behavior recognition technology as, voice because changes of range too big, therefore hard for some precise of match; voice will with volume, and speed and sound of changes (for example Dang you cold Shi) and effect to collection and than on of results; with technology of development, may you can awareness and refused to recording of voice, however, currently for, we also is easy with recorded in tape Shang of voice to cheat voice recognition system. High fidelity microphones are very expensive. Fingerprint recognition system fingerprint recognition as recognition technology has has is long of history has, has solid of market backing, according to General people of views, fingerprint recognition technology through analysis fingerprint of global features and fingerprint of local features, features points as Crest, and Valley and end, and points fork points or differences points, from fingerprint in the extraction of features value can very of detailed to reliable to through fingerprint to confirmed a people of identity.
On average each fingerprint has several unique characteristics that can be measured, each feature has about seven characteristics, our ten fingers to produce the least 4,900 individual measurable characteristics? It was enough to confirm that fingerprint identification is a more reliable method. Here, we generally give the fingerprint image, extract features and processes.
 6, optical technology and capacitance in image acquisition technology
Two kinds of technology used to collect fingerprint images for optical technology and capacitive technology. Optical technology needs a light source from the reflections of prisms in a take like finger light finger to collect fingerprints.
Semiconductor technology with capacitive technology, pressed on to capture finger ridges and valleys at different capacitance between finger skin and chip, chip space, by measuring the different capacitance field full of fingerprints.
Cons: because the chip capacitor expensive chip size and finger are quite expensive, it launched several companies are trying to provide a smaller chip than fingerprints only partially capture the fingerprint to verify using this collection, users must carefully placed on the fingers to ensure a correct reading. And this will inevitably make the read head is difficult to use, another disadvantage of using this chip is only used some of the fingerprints must not collect all fingerprints were compared to the far more reliable.
Another disadvantage of capacitive collection head is susceptible to interference from 60HZ interference to cable users to contact interference, interference within the fingerprint sensor.
Last issue is the reliability of the capacitor collection head, whether it is static, salt from the sweat or other booty and finger-wear will make collecting difficult to read fingerprints.
In fact, so far, optical collecting head provides a more reliable solution. By improving the optical imaging technology, a new generation of optical fingerprint sensor is to impeccable performance and relatively low prices dwarfed the capacitor scheme.
7, fingerprint recognition
Advantages
Fingerprint is human unique of features, and they of complex degrees enough to provides for identification of enough features; if we wants to increased reliability, we just registration more of fingerprint, identification more of finger, up can up to 10 a, and each a fingerprint are is unique of; scan fingerprint of speed soon, using very convenient; read fingerprint Shi, user required will finger and fingerprint collection head mutual contact, and fingerprint collection head directly contact is read human biological features most reliable of method. Fingerprint identification technology is able to occupy most of the market for this is one of the main reasons. Fingerprints can be more compact, and the prices are low.
Disadvantages
Certain people or certain groups of fingerprint minutiae rarely, so it's hard to image; because the use of fingerprints in criminal records in the past, making some people fear "fingerprint record". In practice, however, fingerprint identification technology can now promise not to store any data with fingerprint images, but only store encrypted fingerprint data obtained from the fingerprint. Every time when using fingerprints in the fingerprint left on the user's fingerprint impressions and fingerprint traces are used to copy the possibility of fingerprints. One can see, fingerprint recognition is by far the most convenient and reliable biometric technology, non-infringement and cheap solution for application of the General market has great potential.
Fingerprint recognition technology in the software and firmware despite the importance of price, size and hardware designed for fingerprint identification systems are very important, but that the whole system is robust, but the firmware and software of importance in the overall system status is not lower than the hardware, especially wanted to capture the market for products.
Fingerprint collection head in the of firmware is responsible for processing image and and PC machine of connection, in most of system in the, firmware relative is very simple, it kept of will data transfer to computer, however such do exists with many of problem, first a main of problem is, such transmission data is easy was records and was again using, this makes of system by potential of dangerous, another problem is we required for fingerprint collection head supply power. Computer image phone must keep in order to determine whether there is a fingerprint pressed on it, so that you can at the most appropriate time to capture fingerprint images. By firmware designed to deal with these issues, can improve the performance of the entire system; using the USB interface is currently a good interface solution, can provide power, bandwidth throttling, and plug and play functions fingerprints need to be transferred to the computer the necessary encryption to ensure security, and does not read the fingerprint of the State, the firmware should be transferred to a low power state.
When the host security from your computer after getting the fingerprint image recognition algorithms for further verification process. Fingerprints are so reliable biometrics, that little information can be compared. This characteristic of the fingerprint is not reflected in most fingerprint identification algorithm, on most systems require the user to press all the fingers, fingers on the users have to be very careful. If fingerprints are out of or low quality of fingerprints can cause validation is not possible, you must again press fingers, and such products are not in the market based on.
Therefore, a good system should be more easy to use and reliable, users don't have to worry about placement of the fingerprint algorithm to support a 360-degree rotation and incomplete fingerprints. Users only need to gently press fingers without worrying about whether the right or pressing the part. Finger pressure, rotation, quality, as well as dust and mist collection head, the system has to be a good solution.
Monitoring and surveillance-free operation, one for use with a fingerprint identification technology must first of all understand the monitoring and monitored than the difference between free, in a traditional criminal fingerprint database, fingerprint is obtained in the monitoring State and many times is based on the same database than base. Fingerprint identification algorithm to market to, fingerprint reading must be established in the absence of monitoring State. In this way, just gently press without having to wait for the image to achieve the best, in addition, when dealing with e-commerce, data with the already-registered fingerprint comparison, in most cases is a one-to-one ratio of not more than the right. Monitor-free imaging and one-to-one alignment algorithms and a number of algorithm in the system have a very good strategy, no monitoring mode must be enabled alignment algorithm to deal with poor quality fingerprints and algorithm must be more reliable than the monitor State.
New era of biometric technology
Security requirements bring the accelerating popularity of biometric technology. We believe that the security requirements is a core driver of biometrics market development. 1) personal safety needs: outbreak of the biometric identification technology application in the intelligent terminal market. According to our investigation of the industry chain speculated that after following the application of fingerprint identification technology for the first time in iPhone5S, issued in September, Samsung Note4 phone or will carry an iris scan recognition to improve the safety of mobile phones and other technology vendors are also soon followed suit. 2) public safety needs: Government-led medium and large applications will accelerate. With the promotion of global economic development and urbanization, countries increase investment in critical public safety regulation, such as, in the United States, European Union, India, as the representatives of major Governments around the world have implemented large biometrics system, promote the accelerating popularity of biometric technology in the field of public safety. Global biometrics market over the next five-year compound growth rate up to 35. 2%, the market will reach $37 billion.
According to our model is expected to fingerprint recognition, face recognition, iris recognition "troika" biometric market size by 2018 to 370. US $200 million, 2013-2018 compound annual growth rate of 35. 2%. Structure, fingerprint and iris recognition driven mainly from the outbreak of the intelligent terminal market, face recognition, driven mainly by demand from public safety. Face recognition and iris recognition due to the small base explosive, its share in the biometric market as a whole, from 2013 11. 4%, 5. 1%, up to 2018 in 22. 4% and 22. 1%. China, as a populous country, the need for information security, public safety needs, standardization of biometric drive driven by multiple factors such as, biometric open potential market is accelerating.
Conclusion
Biometric technology is more than just mentioned above, fingerprint recognition, face recognition technology, also include many Palm vein recognition, recognition and other technologies. But fingerprints and face recognition is currently and most widely used technology in the future. In particular, face recognition technology, if we can make good use of already deployed cameras to now, will be able to effectively prevent the occurrence of crime. Of course, privacy issues need to be further explored. It is worth mentioning