Home > News > Industry News > Different biometrics have diff.....
Browse Categories
New Arrival(26)
Hot Products(12)
Reader & Modules Series(20)
Handheld Terminal Series(14)
POS Terminal(5)
Time Attendance Series(14)
Access Control Series(6)
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

Different biometrics have different characteristics in different application areas

  • Author:HFSecurity
  • Release on :2019-11-19

In current research and application fields, biometrics is mainly related to computer vision, image processing and pattern recognition, computer auditory, speech processing, multi-sensor technology, virtual reality, computer graphics, visualization technology, computer-aided design, and intelligence. Other related research such as robot perception systems. Biometrics that have been used for biometric identification include hand shape, fingerprint, face shape, iris, retina, pulse, auricle, etc. Behavioral characteristics include signature, sound, button strength, etc.
Based on these characteristics, biometrics technology has made great progress in the past few years. The biometric identification technologies currently researched and used mainly include face recognition, iris recognition, fingerprint recognition, palmprint recognition, signature recognition, and voice recognition. Different biometric technologies have obvious differences in accuracy, stability, recognition speed and convenience, so they have different characteristics in different application fields.

1、Biometric fingerprint scanner recognition

The fingerprint identification method is the earliest, most extensive and mature biometric identity identification method. At present, the fingerprint identification algorithm is mature, fast, and has good real-time performance. Because of its lowest threshold, low cost and convenient use, it has been widely used in security, attendance, banking, mobile phones and other fields.

The fingerprint identification system is mainly composed of fingerprint image acquisition, preprocessing, feature extraction, comparison, system management and database management. Fingerprint images are collected by optical total reflection photography, silicon crystal sensors, and other techniques (ultrasonic scanning, temperature sensors, pressure sensors). Fingerprint identification methods have their limitations. The fingerprint quality of manual laborers whose fingers have been rubbed for a long time is extremely poor. In addition, there are few people whose fingerprints have very few endpoints and bifurcation points. It is almost impossible to apply fingerprint identification methods for this type of population. .

With the continuous maturity of related devices and algorithms for fingerprint recognition systems, related research focuses on the accuracy of fingerprint images, especially in the case of defects or defaces; on the other hand, it focuses on the continuous reduction of sensor scales. Can be applied to PDAs, mobile phones and other similar devices. For example, Mitsubishi Electric Corporation of Japan has miniaturized the fingerprint authentication device and built it in the mobile phone that the company is about to launch. When the user makes a call, the user can immediately recognize whether the fingerprint is in advance with the user. The registered fingerprints are consistent. If the fingerprint does not match the previously registered fingerprint, the phone cannot be connected. This allows mobile phone users to no longer have to worry about theft of their mobile phones.

2, iris recognition

The iris, as an important identity identification feature, has the advantages of uniqueness, stability, reliability, collectability, and non-invasiveness. And the recognition speed is faster, there is a recognition distance limit, and most of them are used in industrial fields.

The iris recognition system can generally be composed of iris positioning, image preprocessing, iris feature extraction and classification recognition. The working principle of the iris recognition system generally collects the iris image through the iris acquisition device, and then eliminates the noise spots of the image and the influence of the illumination on the image through preprocessing. Using the geometric properties of the iris, by positioning the inner and outer edges of the iris, the upper eyelid, and the lower eyelid. The iris is segmented from the original iris image. The segmented iris image also needs to eliminate the effects of rotation, scale changes, and lash covering caused by the iris image.

The normalized iris image is further enhanced to obtain a preprocessed image to extract corresponding features. In addition, the feature extraction module of the iris is system dependent, as different systems may employ different iris features. After the iris feature is extracted, it is transformed into a pattern recognition problem. The classifier module classifies the characteristics of the iris. If it is an iris identification problem, it only needs to be compared with the target iris. If it is iris recognition, it is necessary to retrieve the corresponding database and obtain the final discrimination result.

Among the various iris recognition algorithms, the iris recognition algorithm based on iris Gabor feature and coarse phase quantization and image registration based on Daugman and Wildes is the most classic. Most commercial systems are based on these two algorithms.

3, face recognition

Face recognition has always been active in our lives. Although the accuracy of face recognition is lower than that of iris and fingerprint, face recognition is the most acceptable biometric method because of its non-invasiveness and the most natural and intuitive way for users. And the collection cost is low, which is a well-recognized identity identification feature, and it is a hot spot in the field of biometric identity identification at home and abroad.

Face recognition has two aspects: positioning the face in the input image; extracting the face feature for matching recognition. In the current face recognition system, the background of the image is usually controllable or approximately controllable, so face positioning is relatively easy to solve. Face recognition, because of the expression, position and direction, is displayed in the form of illumination, which makes it produce larger similar differences, making the feature extraction of the face more difficult.

4, palmprint recognition

Palmprint recognition is more acceptable than fingerprint recognition. Its main features are much more obvious than fingerprints, and it is not easy to be disturbed by noise when extracting. In addition, the necessary features of palm print are more stable and more classified than hand features, so palmprint recognition should be a promising identity. recognition methods.

5, retina recognition

The retina is a very small nerve located at the back of the eye. It is an important organ that the eye perceives light and transmits information to the human brain through the optic nerve. Since the retina is not easily altered and replicated, it is considered the safest way to biometrically recognize it. In the 1930s, people came to the conclusion that the distribution of blood vessels in the posterior part of the human eye was unique. According to further research, even in twins, this blood vessel distribution is unique.

6, signature recognition

Signature recognition is a behavior recognition technology, and most of the signatures are currently only used for authentication. The difficulty of signature authentication is that the dynamic range of data changes, even if the same person's two signatures will never be the same. Signature authentication can be divided into two types according to the way data is obtained: offline authentication and online authentication. Offline authentication is a digital image obtained by a scanner; online authentication is a process of recording a signature using a digital tablet or a pressure sensitive pen, and offline data is easily acquired. However, the dynamic characteristics of the stroke formation process make the online signature easy to be forged.

7, gait recognition

Gait recognition is the analysis of image sequences containing human motion. It usually includes three processes: gait detection, gait characterization and gait recognition. Gait detection is to extract the human gait contour region from the background image in the image sequence. The effective segmentation of the gait contour region is very important for post-processing such as feature extraction and target classification. Therefore, gait detection is often regarded as gait. The pre-processing part of the identification.

Application field

Biometrics technology has been widely used in various fields, and several applications are briefly introduced here.

Aviation identity certification. Airport flights allow passengers to enter through the face, iris and fingerprints. Passengers install a facial camera near the self-service boarding gate, automatically collect their facial images, send the avatar to the internal database for real-time matching and verification. After a few seconds of matching verification, the system identifies the passenger as “boarding” without having to Show your boarding pass or passport at the gate.

Building access is safe. The employees of the enterprise and the owners of the buildings enter and exit the various buildings to carry out the necessary identification to prevent unauthorized access. Currently, various cards, certificates, and the like are used. The problem is that cards and documents are easily lost, copied, and borrowed, and carry troubles.

E-commerce. All companies that send materials over the Internet are looking for a way to send materials securely. More and more people are buying goods and services over the Internet, but the potential security issues are getting worse. It is obviously outdated to replace traditional direct contacts with identity numbers and passwords. The CA certification system is considering incorporating biometrics into it.

ID card forgery. ID cards have always been the object of severe crackdown by state security agencies, but they have been repeatedly banned. The electronic identity card is used to store the physiological characteristics of the person, and the identity card needs to be compared with the biometric characteristics of the person, which will effectively solve the problem of forgery.