In the biometric system, in order to prevent malicious persons from forging and stealing the biometrics of others for identity authentication, the biometric system needs to have a live detection function, that is, to determine whether the submitted biometrics are from a living individual.
General biometric live detection technology uses people’s physiological characteristics. For example, live fingerprint detection can be based on information such as finger temperature, perspiration, and electrical conductivity. Live face detection can be based on head movement, breathing, red-eye effect and other information. , Living iris detection can be based on iris vibration characteristics, eyelash and eyelid movement information, pupil contraction and expansion response characteristics of visible light source intensity, etc.
With the application of deep learning methods and the development of artificial intelligence, the recognition rate of face recognition technology has been qualitatively improved. Through repeated development and experimentation, face recognition technology compares favorably with other biometric recognition technologies. Natural and unique advantage: It can be directly obtained through the camera, and the recognition process is completed in a non-contact manner. Through the comparison of face recognition and certificate recognition, our company's face recognition technology has been applied in the fields of finance, education, scenic spots, travel, social security, etc.
Introduction to Face Recognition Technology
Face recognition technology is mainly divided into two parts:
The first part is the front-end face live detection technology, which mainly supports the android and ios platforms. The front end uses a combination of blinking, opening the mouth, shaking the head, and nodding to ensure that the operation is a real live face.
The second part is the background face recognition technology. This link obtains the entire face image in the live body detection technology link, and then scans the ID card to identify the ID card. After obtaining the ID card portrait, it will compare the face of the scene with the person on the ID card. The face is compared and recognized to determine whether it is the same face.
A face recognition system that can work normally needs other technologies to assist in addition to realizing people recognition. Among them, a very important technology in the face recognition identity authentication system is living body detection.
In addition to “recognizing peopleâ€, the face recognition system also needs to “recognize the truthâ€. In other words, in front of the system, it must not only prove whether the person’s face is the person’s face, but also whether the face is alive. Human faces, not pictures, videos, or masked faces.
Face live detection technology has multiple countermeasures against attacks. Let's explore the mystery of live detection together.
Motion detection
Judge whether the user is operating normally, and by specifying the user to perform random actions (shaking his head, nodding, staring, blinking, and moving the phone up and down) to prevent video attacks and attacks from abnormal actions.
The research of face recognition technology for living body detection still needs a breakthrough in time and space. Whether it is a real person or a photo taken by a camera, the final result is a two-dimensional picture. Therefore, it is difficult to judge whether a real person or a photo is in front of the camera with the current face recognition technology. In addition, face recognition for the recognition of twins, plastic surgery and other groups also needs in-depth research. In the final analysis, face recognition is based on human judgment standards, using deep neural networks and computer technology to extract effective recognition features from face images for identity judgment. When it is difficult for humans to judge with the naked eye, it is still difficult to make a correct identification with current technology and theories.
Near-infrared face detection
The near-infrared face detection is mainly based on the optical flow method.
Near-infrared face detection does not require instruction and cooperation, and the detection success rate is high. According to the optical flow method, the time domain change and correlation of the pixel intensity data in the image sequence are used to determine the "motion" of the respective pixel positions, and the operating information of each pixel is obtained from the image sequence, using Gaussian difference filter and LBP feature Perform statistical analysis of data with support vector machine. At the same time, the optical flow field is more sensitive to the movement of objects, and the optical flow field can be used to uniformly detect eye movement and blinking. This living body detection method can realize blind detection without the user's cooperation.
Silent Living Test
Random action live detection has a high degree of security, but it is more rigid to ask the user to do the action according to the instructions, and the experience is not the best for the user. In response to the increased requirements for user experience, a silent live detection technology has emerged. There is no need for the user to do any action, just need to face the camera naturally for three or four seconds to complete the entire detection process. Compared with photos, real human faces will have micro-expressions even if they do not deliberately perform actions. For example, eyelids, eyeball rhythmic blinking, lips and peripheral cheeks can be effectively prevented by using these features.
Continuity test
Using it together with face detection can better prevent people from switching midway. Verify that the human face movement track is normal. If the person is replaced in the middle, abnormal movement will occur; from a safety perspective, it can prevent skipping the live detection and directly replacing the collected photos.
In addition, in order to prevent tampering with photos taken by living organisms, the collected photos can also be encrypted to ensure information security.
In addition to software, live detection can also be performed through hardware + software. Upgraded configuration of 3D camera + live detection technology to determine whether the face in the image is a real face, prevent image and video attacks, and further enhance the security of the application.
VOZOL BAR 500 Vapes are so convenient, portable, and small volume, you just need to take them
out of your pocket and take a puff, feel the cloud of smoke, and the fragrance of fruit surrounding you. It's so great.
We are China leading manufacturer and supplier of Disposable Vapes puff bars, vozol bar 500 disposable vape,vozol bar 500 vape kit,
vozol bar 500 vape pen, and e-cigarette kit, and we specialize in disposable vapes, e-cigarette vape pens, e-cigarette kits, etc.
vozol bar 500 disposable vape,vozol bar 500 vape kit,vozol bar 500 vape pen,vozol bar 500 vape grape ice,vozol bar 500 Vape blue razz ice
Ningbo Autrends International Trade Co.,Ltd. , https://www.mosvape.com