As an important personal identification method, the automatic face recognition system was first used for criminal photo management and criminal investigation. Now this technology has many applications in security systems and business systems.
Face recognition is directly, friendly, convenient, and robust compared to other methods of identity authentication. Its application fields are gradually pushed to various fields of daily life. On the one hand, it has obviously improved the work efficiency. On the other hand, it also has extremely high safety and reliability, and its application prospect is very broad.
The main use of face recognition technology
In general, face recognition technology has two main purposes in daily life. One is to verify whether you are a certain person or not. This is a 1:1 face verification. That is to say, we first tell the face recognition system, I am Zhang San, and then used to verify whether "I" standing in front of the machine is Zhang San. The second is to let the system automatically identify "who am I." After the system collected one of my photos, I automatically found out who I was from tens of thousands of people, or millions of people.
These are two very different applications for face recognition.
Face recognition technology has developed to the present day, the first use - 1:1 face verification, currently in the controllable environment, has basically reached the point of use; and the second such dynamic recognition, let The system automatically recognizes "who am I", and there is still a long way to go. The current technology has not yet reached the needs of this practical application.
The application of face recognition technology in the financial field
Brushing face cards, remote loans, self-opening accounts, and paying for face-to-face payment... With the rise of face recognition technology in the financial industry, more and more commercial applications have surfaced. Not only ant Jinfu, Weizhong Bank and other emerging Internet financial institutions, traditional financial institutions such as state-owned commercial banks, securities, insurance, etc. have deployed face recognition technology.
Basically, face recognition technology has three main application directions in the financial field: self-service terminals, counter systems, mobile finance and marketing. â€
Face recognition" self-service terminal
To put it simply, the face recognition system is introduced into the self-service device, and the face-receiving technology is used to compare the photos collected on the spot with the stored photos and ID photos and provide similar values ​​for the faces. The staff can be similar. The value of the value is judged whether it passes directly or is manually audited. At present, users can implement self-service card opening, business change, password reset and other personal services on the self-service terminal.
Mobile finance / marketing
The core lies in the remote identity verification of the face, which generally includes two aspects. On the one hand, the user can perform face identity verification by means of a mobile device such as a mobile phone; on the other hand, the financial institution can embed the face recognition system into the portable mobile terminal, and the door is The client handles the business.
Counter system
The core lies in the face network verification. By comparing and verifying the photos of the site with the photos of the existing ID cards of the Ministry of Public Security, it is more objective and scientific to achieve the "humanity and humanity" and reduce the subjective consciousness and mistakes of the "eye" observation. identify. At present, it has been widely used in banking, banking, insurance, securities and other financial institutions to open accounts and other businesses.
Application of Face Recognition Technology in Public Security System
The face recognition photo comparison system is used for rapid identification, in a large number (thousands to millions) of databases (missing population photo library, blocked photo library, resident population photo library, temporary resident photo library, chase Search for the identity of a specific person in the photo library of the escaped person, the photo library of the key population, the CCIC fugitive photo library, etc.
For example, in the provincial, municipal, and other units to establish a comparison center, in the branch office, police station, criminal investigation center to establish a comparison client, or through GPRS / CDMA with mobile phones and PDA photos and send photos to compare requests. It makes full use of the very valuable face photo clues, greatly speeds up the process of identification of suspects by public security investigators, and accelerates the process of “scientific and technical police†to form a highly intelligent, socialized and large-scale public security system. Provide effective technical means.
The face recognition technology focuses on the registration management, the network pursuit, the comparison verification and the post-processing. At the same time, the portrait comparison can also be used for criminal investigation and maintenance of social stability.
The main application directions are as follows:
1. The public security system captures fugitive offenders: The photo comparison system based on face recognition helps speed up the identification of suspects, reduces the inefficiency of “human tacticsâ€, and plays a huge role in applications such as pursuit, crime detection, and tracing. energy.
2, find people to find relatives: photos provided to ordinary people or other business departments, directly into the system for comparison, retrieval, screening, and finally manual confirmation.
3. The police station blocked the illegal personnel: the personnel who were intercepted by the police station, registered transcripts, and those who could not verify their identity, such as some ethnic minorities, deaf or silencers, could take photos and send them to various photo libraries. , to investigate the personnel involved in the major case, so as not to miss the network; or to verify their previous records, cumulative processing.
4, verify the name of the corpse: when you need to verify the unknown corpse, first take a positive photo, send it to the computer, if the photo is closed, damaged or deformed, you can use the portrait synthesis system or manually draw a standard photo, send to the comparison system For verification.
5. Witness description: After obtaining the image description of the suspect by the witnesses on the spot, the portrait synthesis system can be used for investigation.
6, get video surveillance photos: the general monitoring system for the scene, the images of the suspects involved have blurred, deflected, reverse side light and other quality problems, then you need to use the portrait synthesis system or manually draw a standard according to the image Photographs are sent to the photo comparison system for verification.
7. Public meeting gatherings: In public places such as the government and stadiums, there are always people who are in trouble. At this time, the public security police can not directly handle people. You can use a telephoto camera to take close-up shots. If the effect is not good enough, you can use the portrait synthesis system to correct it. The comparison is checked by the comparison system.
8. Identification of first-generation/second-generation resident ID cards: According to the photo information of the ID card of the offender, compare it with the information in the photo library of the system, and extract the information of the person similar to the photo on the certificate, so that the existing two can be fully utilized. The ID card photo resource provides efficient and beneficial help for the work of the public security department.
9. Missing population surveys, shelters, compulsory detoxification, customs entry and exit, etc. The photo comparison system helps improve work efficiency and greatly reduces the intensity of manual comparison.
10. Other applications: comparison of resident population, comparison of temporary population, comparison of key populations, comparison of CCIC fugitives, etc.
Intelligent face recognition analysis technology to achieve accurate calculation and retrieval of the tens of millions of photo libraries, screening of the required personnel information, and to ensure that the results of the analysis can meet the requirements of manual secondary analysis and processing, face recognition system implementation architecture The picture is as follows:
Application of face recognition technology in social security field
In order to ensure the safety of the basic endowment insurance fund, prevent fraudulent use of pensions, and introduce face recognition technology into face recognition self-certification (personal authentication system), the uniqueness of face recognition features is used to achieve accurate and convenient Effective verification of the identity of the insured person, reducing the loopholes in the insurance, and avoiding the problems of fraudulent insurance and fraud.
The self-certification terminal automatically reads the information in the document locally, and automatically detects and captures the face of the verified person, obtains the face photo on the document, passes the high-definition face recognition analyzer, and the on-site certificate holder Faces are compared to check whether the holder is the same person as the current one, thus effectively preventing all fraudulent behaviors in the authentication work, including fraudulent use of photos, videos and 3D models for authentication.
System structure diagram:
The application of face recognition technology can not only effectively stop the loss of pensions, greatly reduce the workload and improve work efficiency; at the same time, it can optimize the quality of service services and reduce service friction, which greatly facilitates the retirees and truly plays a role. Three arrows, one stone and three birds.
Application of Face Recognition Technology in Airport Border Inspection
With the accelerated pace of urban life and the continuous improvement of living standards, the aircraft has become a major travel choice besides trains and cars. However, the current airport check-in procedures and security measures are complex and weak.
For public places such as airports, where security is intensive, security has always been a top priority for airport management. Many airports have begun to use the HD face document comparison system to assist the airport in manual inspection.
The core of the HD Face Document Matching System is the comparison of ID cards and portraits. Quickly identify whether the certificate is consistent with the user of the certificate, and the recognition rate is over 98%, which is faster and more accurate than the naked eye. When the document information is found to be inconsistent with the license holder, the system will automatically prompt the security personnel to strengthen the manual verification work.
When the passenger is ready to enter the waiting hall, the camera at the security checkpoint will automatically capture the face image, and the face recognition system will automatically compare the passenger ID photo with it to identify the passenger. Even if the passenger changes his hair style and makes heavy makeup, it doesn't matter. The face image collected by the face recognition system can also be recorded as very important monitoring data, stored in the database, used as an index for post-mortem retrieval, or connected with the database of the public security and security departments for forensics and identification.
In the airport entry and exit security detection system, the face recognition system plays multiple functions. The first one can control and control the traffic flow of the airport, and the second can verify and test the identity information of the entry and exit personnel, and do not give suspicious persons the opportunity to enter or flee.
Technical difficulties in face recognition
Although face recognition has been developed for 3 or 40 years now, it has always had several difficulties, and it has not been completely solved until now.
1, lighting problems
The illumination problem is an old problem of machine vision, especially in face recognition. Due to the 3D structure of the face, the shadow projected by the light will strengthen or weaken the original facial features.
2, expression gesture problem
Similar to the illumination problem, the pose problem is also a technical difficulty that needs to be solved in the face recognition research. The pose problem involves facial changes caused by the rotation of the head about three axes in a three-dimensional vertical coordinate system, where depth rotation in two directions perpendicular to the image plane causes partial loss of facial information. The research on pose is relatively rare. At present, most face recognition algorithms mainly focus on frontal and quasi-positive face images. When the pitch or left and right sides are more severe, the recognition rate of the face recognition algorithm will also be Will drop sharply. Facial expressions such as crying, laughing, and anger that have a large facial amplitude also reflect the accuracy of facial recognition.
3, occlusion problems
For face image acquisition in non-cooperating situations, the occlusion problem is a very serious problem. Especially in the monitoring environment, often the monitoring object will bring glasses, hats and other accessories, so that the collected face images may be incomplete, which affects the subsequent feature extraction and recognition, and even leads to face detection algorithm. Failure.
4, age changes
As the age changes, the appearance of the face also changes, especially for adolescents, this change is more obvious. The recognition rate of face recognition algorithms is different for different age groups. When a person turns from a teenager to a youth and becomes an old person, his appearance may undergo a relatively large change, resulting in a decline in the recognition rate. The recognition rate of face recognition algorithms is different for different age groups.
5, face similarity
The difference between different individuals is small, the structure of all faces is similar, and even the structural shapes of face organs are similar. Such a feature is advantageous for positioning with a human face, but is disadvantageous for distinguishing human subjects with human faces.
6, image quality
The source of face images may be varied. Due to the different acquisition devices, the quality of face images obtained is different, especially for those with low resolution, high noise and poor quality (such as those photographed by mobile phone cameras). Face images, pictures taken by remote monitoring, etc. How to perform effective face recognition is a problem that needs attention. Similarly, the impact of high-resolution images on face recognition algorithms needs further research.
7, lack of sample
The face recognition algorithm based on statistical learning is the mainstream algorithm in the field of face recognition, but the statistical learning method requires a lot of training. Since the distribution of face images in high-dimensional space is an irregular manifold distribution, the available samples are only samples of a very small part of the face image space. How to solve the statistical learning problem under small samples needs further Research.
8, massive data
Traditional face recognition methods such as PCA, LDA, etc. can be easily trained in small-scale data. But for massive data, these methods are difficult to train and may even collapse.
9, large-scale face recognition
As the size of the face database grows, the performance of the face algorithm will decline.
Storage Battery,Auto Battery,Car Battery,Autocraft Battery
SUZHOU DEVELPOWER ENERGY EQUIPMENT CO.,LTD , https://www.fisoph-power.com