The development of technology is accelerating the transformation of our lives. In the past, when we were shopping for a bill, the cashier would ask “cash or credit card.†Now, this sentence becomes “WeChat or Alipay?†Before we went to the street to bring cash, it later became a card, now only With shou machine.
However, have you thought about it? One day in the future, we will not even have to bring the shou machine to the streets, as long as "face". Because, we are moving towards the "brushing era." At that time, all your information and property will be bound to your face, and you can go out and "brush your face". Today, let's take a closer look at face recognition technology:
Face recognition overview
Face recognition is a biometric recognition technology based on human facial feature information for identification. A series of related techniques for capturing a face or a video stream with a camera or a camera and automatically detecting and tracking the face in the image, and then performing face detection on the detected face, which is also commonly called portrait recognition and face recognition.
Face recognition is a popular field of computer technology research. It belongs to biometrics technology, which is a biological feature of an organism (generally referred to as a person) to distinguish organisms.
Biometrics studied by biometrics include face, fingerprint, palmprint, iris, retina, sound (speech), body shape, personal habits (such as the strength and frequency of typing on the keyboard, signature).
Corresponding recognition technologies include face recognition, fingerprint recognition, palmprint recognition, iris recognition, retina recognition, and speech recognition. (Speech recognition can be used for identification, and voice content can be recognized. Only the former belongs to biometric recognition technology). Body shape recognition, keyboard stroke recognition, signature recognition, etc.
Three key technologies
1. Feature-based face detection technology
Face detection is performed by using colors, contours, textures, structures, or histogram features.
2, based on template matching face detection technology
The face template is extracted from the database, and then a certain template matching strategy is adopted to match the captured face image with the image extracted from the template library, and the face size and the position information are determined by the correlation level and the matched template size.
3. Statistical face detection technology
By collecting a large number of face positive and negative sample banks for the "face" and "non-human face" images, the statistical method is used to strengthen the training of the system, so as to detect and classify the face and non-face modes.
Four characteristics
1, geometric features
From the distance and ratio between the face points as features, the recognition speed is fast, the memory requirement is relatively small, and the sensitivity to light is reduced.
2, based on model features
The facial image features are extracted according to the different probabilities of different feature states.
3, based on statistical characteristics
The face image is regarded as a random vector, and different face feature patterns are distinguished by statistical methods, such as typical feature face, independent component analysis, and singular value decomposition.
4, based on neural network characteristics
The face image features are stored and memorized by a large number of neural units, and the face images are accurately identified according to the probability of different neural unit states.
Ten difficulties
1, lighting problems
The change of illumination is the most important factor affecting the performance of face recognition. The degree of resolution of this problem is related to the success or failure of the process of face recognition. Due to the 3D structure of the face, the shadow projected by the light will strengthen or weaken the original facial features. Especially at night, facial shadows caused by insufficient light can cause a sharp drop in the recognition rate, making it difficult for the system to meet practical requirements.
At the same time, the theory and experiment also prove that the difference between the same individual due to different illumination is greater than the difference between different individuals under the same illumination. Illumination is an old problem in machine vision, especially in face recognition. Solutions to solve lighting problems include 3D image face recognition and thermal imaging face recognition. However, these two technologies are still far from mature and the recognition effect is not satisfactory.
2, posture problem
Face recognition is mainly based on the facial features of the person. How to recognize the facial changes caused by the gesture becomes one of the difficulties of the technology. 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. Make the attitude problem a technical problem of face recognition.
The research on attitude 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 A sharp decline.
3, expression problems
Facial expressions such as crying, laughing, and anger that have a large facial amplitude also reflect the accuracy of facial recognition. The existing technology handles these aspects well, whether it is opening mouth or doing some exaggerated expressions, and the computer can correct it through three-dimensional modeling and gesture expression correction.
4, 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 monitored objects will wear 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 algorithms. Failure.
5, age changes
As a person changes from a teenager to a youth and becomes an old age, 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.
The most direct example of this problem is the identification of ID card photos. The validity period of ID cards in China is generally 20 years. In these 20 years, the appearance of each person will inevitably undergo considerable changes, and all of them also have great recognition. problem.
6, 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.
7, dynamic identification
In the case of non-composite face recognition, motion resulting in blurred facial images or incorrect camera focus can seriously affect the success rate of facial recognition. This difficulty is evident in the use of security and surveillance identification in subways, highway bayonet, station bayonet, supermarket reversal, border inspection, etc.
8, face security
The main deception of forging face images for recognition is to create a three-dimensional model, or to graft some expressions. With the improvement of face anti-counterfeiting technology, the introduction of intelligent computing vision technology such as 3D facial recognition technology and camera, the success rate of forging facial images for recognition is greatly reduced.
9, image quality problems
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 face images with low resolution, high noise and poor quality (such as shou camera). 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. Now, when we use face recognition, we generally use face images of the same size and close in definition, so the image quality problem can be basically solved, but in the face of more complicated problems in reality, we need to continue to optimize the processing.
10, the sample is lacking
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.
In addition, the face image database that is currently involved in training is basically an image of a foreigner. There are few face images of Chinese and Asian people, which makes it difficult to train the face recognition model.
Face recognition application dimension
1, dynamic scene two dimensions
First, the definition of 1:1.1:1 is a judgmental function. The application scenario is actually in the financial and personal certification, and the characteristics are more precise and safe. So now everyone is Alipay or the bank’s personal certificate. The real name business basically uses the recognition of 1:1 face.
Second, 1:N. 1:N is more in a database or a bottom library, can find the person who is in the bottom library, so it is a process of identification, is a dynamic, or a non-matching scene.
For example, in security, I went to pick up fugitives. When I went to catch fugitives, I couldn’t let fugitives see the camera. In the business scenario, it is impossible for our VIP customers, employees, and members to do the operation of the camera head, so it is a dynamic and non-matching scene.
2. Four dimensions of the business scenario
First, the plates are large enough to support the company's long-term development.
Second, the data is backflowing.
Third, is it the use of high frequency scenes and high frequencies.
Fourth, whether it can be copied, can be changed from 1+0 to 1+N to improve efficiency.
3, three dimensions of the visualization system
First, personnel management.
Second, the fusion of sensor networks.
Third, the integration of commercial real estate + new retail as a whole.
Face recognition application field
Application areas: finance, justice, security, border inspection, aerospace, electricity, education, medical, etc.
Four potentials of commercialization: gates, transportation, banks, shou machines
Specific application scenario overview
I. Financial field
1. Face recognition autonomous terminal
Manual review, self-opening, business change, password reset, and other personal business.
2. Mobile finance and sales
Remote authentication verification, two aspects: verification of user identity and portable devices with face recognition systems required for financial institutions to conduct business on their door.
3, counter system
Face network verification, used for bank accounts, insurance, securities and other financial institutions to open accounts, and other services.
Second, domestic airport applications
Three key points: first attempt, boarding, and comprehensive intelligence
Iconic events:
1. The first attempt of Beijing Capital Airport in 2009 was the first step for domestic airports to begin to recognize face recognition technology. However, limited to the level of face recognition technology at the time, they had to use magnetic cards for cross-validation to ensure the uniqueness of identity. In the recognition speed and accuracy, the face recognition technology at that time and the face recognition technology after deep learning intervention are not at a level.
2. In 2014, Nanjing Lukou Airport first tried to apply face recognition technology to boarding. Although it was also limited by the level of technical business at that time, it could not achieve self-service clearance, but it provided a reference for the next application. And experience.
3. In December 2016, the comprehensive intelligence of Yinchuan Airport marked the entry of a new level of airport intelligence. In addition to security clearance and self-check-in, face recognition and related computer vision technology is applied to dynamic control, flow guidance, smart navigation, VIP welcome, track retrieval, cleaning reminder, etc. for face recognition in 2017 Technology has laid a good foundation for the explosion of airport applications.
China Southern Airlines----the first airline in China to use face recognition technology
The CZ3384 became the first flight to board a new technology. Passengers do not need to hold a boarding pass, and they can quickly pass through the boarding gate.
Third, the Chinese style crosses the road
1 Use face recognition to solve cost problems.
2 Adhere to the administration according to law and prevent extrajudicial punishment.
3 Solve the conflict of road rights and avoid law enforcement.
Example: According to the Jinan police, the face recognition system is mainly used to capture pedestrians and non-motorized drivers of red lights, and can also clearly image at night. Pedestrians are “grabbed in the currentâ€, and the short video of the red light and the enlarged avatar will be directly exposed on the display of the intersection and presented to the public.
In addition, the device is also connected to the resident identity information system, and the personal information such as the name of the offender and the ID card identified by the face will also be displayed on the electronic screen.
After the face recognition system was launched in Jinan, more than 6,200 pedestrian and non-motor vehicle red light violations were captured in one month. With the deterrence of "black technology", the behavior of red light has been effectively curbed, and the number of people who have red light every day has dropped from more than 100 times to more than a dozen times.
In Jiangbei, Chongqing, since the face recognition system was put into trial operation, the pedestrian compliance rate has increased from 60% to over 97%.
Hidden dangers: Personal information is publicly available, involving the disclosure of personal privacy issues. Experts suggest that for face recognition and other information gathering activities, it is necessary to make an announcement to the public in advance to inform the public that it has entered the public information collection area, and the illegal acts will be filmed and exposed, thus satisfying the people's right to know and alerting. Role; for the collected information, appropriate technical treatment should be carried out, and privacy that should not be disclosed should be obscured or not disclosed.
Fundamental: The experts interviewed said that the difficulty of crossing the road due to the unreasonable setting of transportation facilities is often the main reason for pedestrians to red light. Some urban road network planning is unreasonable, focusing on the construction of trunk roads, the density of branch roads and secondary roads is not up to the requirements, leading to pedestrians and non-motor vehicles being gathered on the main road; some intersection traffic lights are not allocated reasonably, if If you cross the road according to the rules, you must have enough patience and fast enough speed. Only by comprehensively managing and solving the conflict between the people and the car “road rights†can we fundamentally solve the “Chinese style crossing the roadâ€.
Fourth, the field of education
Candidate identification and identification, campus, dormitory access management and other scenarios.
Example: In the 2016 college entrance examination, Beijing, Sichuan, Hubei, Guangdong, Liaoning, Inner Mongolia and other provinces have adopted the “face recognition + fingerprint recognition†biometric technology to confirm the identity of candidates, preventing the occurrence of test and cheating.
With the expansion of the pilot area and various fields and the maturity of the operation model, the Industry Reporting Institute expects face recognition in 2017 to be popularized.
V. Public security field
(1) Face capture and tracking function.
(2) Face recognition calculation.
(3) Modeling and retrieval of faces.
The use of face recognition products in the public security field is mainly reflected in two aspects, on the one hand, the use of the background dynamic face recognition system, and on the other hand, the use of front-end face recognition handheld devices and personal identification machines.
Sixth, the medical field
1. Community medical examination application
When using digital medical examination equipment (electronic sphygmomanometer, human scale, blood glucose meter, etc.) in the community, the data is transmitted to the electronic medical record or health file, and the living face information of the patient is stored and stored. Identification. Each record after the completion of the unique identification will be recorded, so that the patient's condition can be quickly reported back to the doctor and the patient himself, and the best treatment plan can be easily obtained.
2. Application of secondary and above medical institutions
By setting a face recognition system in different scenes such as a kiosk, a window, a clinic, and the like, the recognized face information is used as an information retrieval portal, the patient's information file is associated, and then the face can be retrieved for medical treatment records, and the like.
Seven, smart city area
1. Pension management
The use of face recognition technology can effectively check personnel and reduce the loss of pensions.
2. Tax certification system
Through the face recognition technology, the system automatically compares the portrait of the lens ingested with the portrait in the identity information of the public security department, and completes the real-name authentication in real time. Not only has it effectively alleviated the pressure on taxpayers in the window, it has improved the efficiency of taxation, and it has also enhanced the experience of real-name taxation and reduced tax-related risks.
3. Suspect tracking system
Based on the face recognition technology, the public places such as long-distance passenger stations and railway stations are monitored, and the faces in the video are compared with the database of suspects. Once the suspects are identified in the crowd, they are immediately alerted. This greatly reduces the workload of managers, improves the efficiency of arrest, and increases the safety of the city.
4. Community Management System
In smart cities, taking the smallest unit community in the city as an example, through non-matching face recognition, the property management department can help the owners to be more friendly in terms of visitor management, property notification (water and electricity fee notification, garage information, etc.). Natural life experience.
5, building access control system
The face recognition intelligent access control system can improve the security of buildings and homes by constructing an intelligent identification management system and combining advanced face recognition algorithms to accurately and quickly identify faces and open access control.
6, candidate authentication management system
Based on the special needs of the examination industry, it integrates computer, communication, network, face recognition technology, database and other diversified technologies into one application system project, providing test institutions with functions such as information extraction, identity verification and management of candidates, and builds more An efficient and fair test environment.
7. Identification information and safe driving management system for driving students
Including on-site verification, student identification, boarding and pick-up, driving time control, etc.
8, intelligent meal management system
The system performs face recognition when the students are cooking, records the dishes that the students eat every day, and then draws dietary adjustment opinions according to the comparative analysis of the hospital physical examination results. It records the foods when the students waste a single amount exceeding the prescribed value, and then continuously optimizes the dishes. To achieve the purpose of adjusting and optimizing the student's diet structure.
9. Business Intelligence Analysis System
The face recognition system can make full use of machine vision to identify and induct the features of the face, and use the gender, age, mood, etc. of the customer as the corresponding features of the business demand, and push the content of interest to the customer in real time to target the merchant. Customer group diversion and accurate sales; on the other hand, through the observation and learning of interest content of different groups of people, gradually improve the matching accuracy of the target crowd push content.
Unclear about face recognition
1. Face recognition is mainly composed of two links:
One is the face comparison, that is, whether the face to be verified is the person or not, and the second is the living body detection, that is, whether the face to be verified is true or not.
2, the way to crack:
The former is very simple to crack, there is a photo of myself, if this person likes to take a selfie, it is too easy, even if you don't love to send, it is not complicated to take a photo.
The second point of living detection is originally the most important part, but with a very simple function, it is cracked. This function was originally used on some beauty cameras. You can put a cute cat-like beard on your face. Now you can put a photo of your head on another person's face, and you can pass the second living body. Detected.
In this way, the entire face recognition is also ineffective. A high-tech smart function has become an unsafe trap.
Yan Shuicheng, 360 chief scientist and dean of the Institute of Artificial Intelligence, said that with the gradual application of face recognition to medical insurance, social security benefits, railway facilities and airport security, major manufacturers still need people at this stage. Face recognition is guarded against the security risks caused by artificial intelligence 'fast running'.
Dean Yan’s suggestion is to combine face and voiceprint, fingerprint, iris and other biometric authentication signals when using high-level security scenarios such as privacy and payment, rather than using face recognition alone. Technology, so the coefficient of safety will be greatly improved.
In short, whether face recognition is safe is only a technical problem, and the security issues in the field of artificial intelligence and other technologies are taken seriously, which is the most important place.
Glazed Tile And Ridge Cap Roll Forming
Machine
Working principle:
steel strip is made into glazed tile by
machine through cold-bend forming technology.
Usage:
Glazed tile produced by roll forming
machine are widely used in construction, transportation and other industries.
It is mainly used for roofing and wall of buildings, such as steel structures,
airports, storehouses,etc.
Component:
1.Decoiler
2.Roll Forming System
3.Cutting Device
4.Hydraulic Station
5.PLC control system
Working Flow:
Loading coil – decoiling – guide feeding –
main roll forming – hydrarlic cutting – finished products
Corrugated Roof Roll Forming Machine Technique
Parameters:
Processing material: aluzinc/galvanized steel/colored steel coil
Material thickness: 0.3-0.8mm
Main motor power: 4kw(can change as request)
Pump power: 4kw
Shaft diameter: 75mm and solid shaft
Forming steps: 15steps and more
Roller material: high grade 45# steel, hard Chromium plating
Main frame:300H steel
Medium plate thickness: 16mm
Speed:8-12m/min
Material of the cutting blade:Cr12
Control system: PLC computer control
Power supply:
380V, 3 Phase, 60Hz (can change as customers request)
Glazed Tile profile series:
Our Glazed Tile And Ridge Cap Roll Forming
Machine can greatly simplifie the production process and improve production
efficiency.
Welcome visit our Factory!