Face recognition technology is based on human facial features, input facial images or video streams. First, it is judged whether or not there is a human face, and if there is a human face, the position and size of each face and the position information of each main facial organ are further given. Based on this information, the identity features contained in each face are further extracted and compared with known faces to identify the identity of each face.
Analysis of the principle of face recognition technology
Faces are attracted by many industry customers due to their easy-to-collect characteristics, especially in public security, customs, and shopping malls. Humans are performing face recognition every day, so they are best able to accept this type of authentication. The study of face recognition began in the middle of the last century. After decades of hard work, it can now be applied in our real life and provide us with various conveniences.
Face recognition is mainly divided into three processes: face detecTIon, feature extracTIon and face recogniTIon.
Face detection: Face detection refers to detecting and extracting face images from the input image. Usually, the haar feature and the Adaboost algorithm are used to train the cascade classifier to classify each block in the image. If a rectangular area passes through the cascade classifier, it is discriminated as a face image.
Feature extraction: Feature extraction refers to the representation of face information by some numbers, which are the features we want to extract. Common facial features fall into two categories, one is geometric features and the other is characterized. Geometric features are geometric relationships between facial features such as the eyes, nose, and mouth, such as distance, area, and angle. Since the algorithm utilizes some intuitive features, the amount of computation is small. However, because its required feature points cannot be precisely selected, its application range is limited. In addition, when the illumination changes, the face has foreign objects obscured, and the facial expression changes, the feature changes greatly. Therefore, this type of algorithm is only suitable for rough recognition of face images and cannot be applied in practice.
The characterization feature utilizes the gradation information of the face image to extract global or local features through some algorithms. One of the more commonly used feature extraction algorithms is the LBP algorithm. The LBP method first divides the image into several regions, and thresholds the center value in the pixel 640x960 neighborhood of each region, and regards the result as a binary number. Figure 3 shows an LBP operator. The LBP operator is characterized by a constant monotonic grayscale change. Each region obtains a set of histograms by such operations, and then joins all the histograms to form a large histogram and performs histogram matching calculation for classification.
Face recognition: The face recognition mentioned here is a narrow face recognition, which compares the features extracted by the face to be recognized with the features of the face in the database, and discriminates the classification according to the similarity. Face recognition can be divided into two major categories: one is confirmation, which is the process of comparing the face image with the existing image of the person in the database, answering whether you are your problem; the other is identifying This is the process of matching the face image with all the images already stored in the database, answering who you are. Obviously, face recognition is more difficult than face recognition because identification requires a large amount of data matching. Commonly used classifiers include nearest neighbor classifiers, support vector machines, and the like.
Similar to the fingerprint application method, the face recognition technology is relatively mature and is also an attendance machine. Because in the attendance system, the user is actively cooperating, and the face that meets the requirements can be obtained in a specific environment. This provides a good input source for face recognition and often results in satisfactory results. However, in some public places, video surveillance probes, due to light and angle problems, the resulting face images are difficult to compare successfully. This is also one of the problems that must be solved in the future development of face recognition technology.
At present, some institutions and universities are conducting research on new areas of face recognition and new technologies. Such as long-distance face recognition technology, 3D face recognition technology. Long-distance face recognition systems face two major difficulties. The first is how to get a face image from a distance. Second, how to identify the identity if the resulting data is not ideal. In a sense, long-distance face recognition is not a specific key technology or basic research problem. It can be seen as an application and system design issue. There are usually two types of workarounds for getting face images. One is a high-definition fixed camera, and the other is a PTZ control system multi-camera system. The latter is more suitable for the general situation, but its structure is more complicated and the cost is more expensive. The latter needs to consider how to coordinate the synchronization of multiple cameras. Typically, the system consists of a low resolution wide angle camera and a high resolution telephoto camera. The former is used to detect and track targets, while the latter is used for face image acquisition and recognition. At present, the long-distance face recognition technology is still in the laboratory stage. If the above problems can be solved in the future, it is of great significance for the application of personnel control.
3D face recognition can well overcome the problems of gestures, illumination, and expressions encountered by 2D face recognition. The main reason is that 2D images do not represent depth information well. Generally, the 3D face recognition method acquires a 3D face using a 3D scanning technique, and then creates a 3D face model and uses it for recognition. However, the shortcomings of 3D face recognition technology are also obvious. First, it requires an additional 3D acquisition device or binocular stereo vision technology. Second, the modeling process requires a large amount of computation. It is believed that with the development of chip technology in the future, 3D face recognition will become one of the hot technologies when computing power is no longer constrained and the cost of acquisition equipment drops dramatically.
Face recognition technology application prospects
As technology matures and social recognition increases, face recognition technology will be applied in more fields.
Challenges in face recognition applications
1. Enterprise, residential security and management. Such as face recognition access control attendance system, face recognition security door and so on.
2. Electronic passport and ID card. This may be the largest application in the future. The International Civil Aviation Organization (ICAO) has determined that from 2010, its 118 member countries and regions must use machine readable passports. Face recognition technology is the first recognition mode. Become an international standard. China's ePassport Program Ministry of Public Security is stepping up planning and implementation.
3. Public security, justice and criminal investigation. Such as the use of face recognition systems and networks, the search for fugitives nationwide.
4. Self-service. For example, if the bank card and password are stolen, the cash will be taken by others. This can be avoided if face recognition is applied at the same time.
5. Information security. Such as computer login, e-government and e-commerce. In e-commerce, transactions are all done online, and many approval processes in e-government have also moved to the Internet. At present, the authorization of transaction or approval is realized by password. If the password is stolen, security cannot be guaranteed. However, using biometrics, the digital identity and real identity of the parties on the Internet can be unified, thereby greatly increasing the reliability of e-commerce and e-government systems.
Electronic Vape E Bottle,E-Cigarette Oil Bottle,E-Cigarette Disposable Vape Oil Bottle,Portable Electronic Cigarette Oil Bottles
Shenzhen Niimoo Innovative Technology Co., Ltd , https://www.niimootech.com