I. Face recognition technology
Face recognition is based on the human face characteristics, to the input face image or video stream . Firstly, it determines whether there is a human face, and if there is a human face, it further gives the position, size and location information of each major facial organ of each face. Based on this information, the identity features contained in each face are further extracted and compared with the known faces to identify each face.
II. Algorithm of face recognition
It integrates computer image processing technology and biostatistics principle in one, using computer image processing technology to extract human image feature points from video, and using the principle of biostatistics to analyze and establish mathematical model, i.e. face feature template. The completed face feature template is used to analyze the features with the human face of the subject, and a similarity value is given according to the result of the analysis. By this value, it can be determined whether it is the same person or not.
III. Main functional modules of face recognition
1、Face capturing and tracking function.
Face capture refers to detecting a portrait in one frame of an image or video stream and separating the portrait from the background and saving it automatically. Face tracking refers to the use of portrait capture technology to automatically track the specified portrait when it moves within the range of the camera shot.
2、Face recognition comparison.
Face recognition is divided into verification type and search type two kinds of comparison mode. The verification type refers to the capture of the human image or the specified human image and the database has been registered in a pair of images for comparison and verification to determine whether it is the same person. The search type of comparison means searching for the existence of a specified person from all the registered images in the database.
3、Modeling and retrieval of faces.
The registered image data can be modeled to extract the features of faces and generate face templates (face feature files) to be saved in the database. In the face search (search type), the specified portrait will be modeled, and then compared with the template of all people in the database for identification, and finally the list of the most similar people will be listed according to the compared similarity value.
4、Real person identification function.
The system can identify whether the person in front of the camera is a real person or a picture. In order to prevent users from using photos to fake. This technology requires the user to make facial expressions with the action.
5、Image quality detection.
The image quality directly affects the recognition effect, the image quality detection function can evaluate the image quality of the photos to be compared, and give the corresponding suggested value to assist the recognition.
Ⅳ. Face Recognition Advantages
The advantage of face recognition lies in its naturalness and the feature of not being detected by the measured individual.
By naturalness, it means that the recognition method is the same as the biometric features used by humans (or even other organisms) for individual recognition. For example, face recognition is used by human beings to distinguish and confirm their identity by comparing faces, and there are other natural recognition methods such as voice recognition and body shape recognition.
Being undetectable is also important for a recognition method to be inoffensive and less likely to be deceptive because it does not attract attention. Face recognition has this aspect characteristic, it completely uses the visible light to obtain the face image information, and is different from the fingerprint recognition or iris recognition, need to use the electronic pressure sensor to collect the fingerprint, or use the infrared ray to collect the iris image, these special collection way is easy to be perceived, thus more likely to be disguised deception.
V. Application products of face recognition in access control and time and attendance system
Huifan FR05 face recognition machine series products, is the world's first embedded "dual-camera" face recognition products, it uses the latest face recognition algorithm, the application of embedded high speed digital signal processing technology, recognition speed, high accuracy, not affected by the environment light;, specifically for the majority of enterprises and institutions access control management and Attendance management and design, safe, convenient, fast.
1. unique face key area positioning method, high positioning accuracy, high recognition rate.
2. face recognition algorithm with self-learning function, which can effectively solve the problem of slow age change of human face.
3. the adoption of special dual cameras, which belongs to quasi-3D face recognition technology, with recognition performance greatly exceeding 2D face recognition and algorithm complexity much lower than 3D face recognition, with obvious technical advantages.
4. The recognition performance is not affected by the ambient light and can be recognized even in the dark, meeting different user needs and usage environments
Ⅵ. the application of facial time and attendance machine
1、Face recognition time and attendance at construction sites
Face recognition technology, as a means of real name verification, fits perfectly with the real name management of construction sites. By entering workers' information into the face recognition system and filing it in the public security system, it can fully avoid the loopholes of losing punching credentials or impersonating punching cards, and at the same time properly manage people's access, solving two major problems of attendance and personnel flow management at one go.
Construction workers only need to enter their own face and identity information in advance, and when clocking in and out of work through dynamic capture of site photos, and the initial face library recorded for comparison, screening out the corresponding personnel information, can be recorded in the attendance information, but also to restrict the outsiders who have not entered the information to enter the site at will.
2、Face recognition time and attendance of office buildings
Face recognition attendance application can also be installed in and out of office premises, and attendance time can be set according to the actual needs of the scene and attendance rules, personnel "brush face" attendance, and the system records personnel attendance accurately in real time. Managers can check the attendance status of personnel at any time through the management background, and the data is clear at a glance and easy to check.
3、Campus Face Recognition Attendance
Unlike traditional paper sign-in, verbal roll call and manual card swiping, students only need to capture face information through the camera to achieve millisecond punching and signing in, and multiple students can be accurately identified through the camera at the same time to achieve batch punching, which greatly saves time for signing in and improves classroom teaching efficiency.
In addition, face attendance makes the sign-in more standardized and avoids the appearance of substitute sign-in. The traceability of face sign-in data makes class attendance and individual attendance at a glance, helping teachers and parents to better understand students' status.