Principles of facial recognition technology
2021-09-30
Technical process
The facial recognition system mainly consists of four components, namely: facial image acquisition and detection, facial image preprocessing, facial image feature extraction, and matching and recognition.
Facial image acquisition and detection
Facial image acquisition: Different facial images can be captured through camera lenses, such as static images, dynamic images, different positions, and different expressions, which can be well collected. When the user is within the shooting range of the collection device, the collection device will automatically search for and capture the user's facial image.
Face detection: In practice, face detection is mainly used for preprocessing of face recognition, which accurately calibrates the position and size of the face in the image. Facial images contain rich pattern features, such as histogram features, color features, template features, structural features, and Haar features. Face detection is the process of picking out useful information and utilizing these features to achieve face detection.
The mainstream face detection methods use the Adaboost learning algorithm based on the above features. Adaboost algorithm is a classification method that combines some weaker classification methods to create a new and strong classification method.
In the face detection process, the Adaboost algorithm is used to select some rectangular features (weak classifiers) that best represent the face. The weak classifier is constructed into a strong classifier through weighted voting, and then the trained strong classifiers are concatenated to form a cascaded hierarchical classifier, effectively improving the detection speed of the classifier.
Facial image preprocessing
Facial image preprocessing: The preprocessing of facial images is based on the results of facial detection, processing the images and ultimately serving the process of feature extraction. The original images obtained by the system are often not directly usable due to various conditions and random interference, and must be preprocessed in the early stages of image processing, such as grayscale correction and noise filtering. For face images, the preprocessing process mainly includes light compensation, gray transformation, Histogram equalization, normalization, geometric correction, filtering and sharpening of face images.
Facial Image Feature Extraction
Facial image feature extraction: The features that can be used by facial recognition systems are usually divided into visual features, pixel statistical features, facial image transformation coefficient features, facial image algebraic features, etc. Facial feature extraction is aimed at certain features of the face. Facial feature extraction, also known as facial representation, is the process of modeling facial features. The methods of facial feature extraction can be classified into two categories: one is knowledge-based representation methods; Another method is based on algebraic features or statistical learning for representation.
Knowledge based representation methods mainly obtain feature data that is helpful for facial classification based on the shape description of facial organs and the distance characteristics between them. The feature components usually include Euclidean distance, curvature, and angle between feature points. The face is composed of parts such as the eyes, nose, mouth, chin, etc. The geometric description of these parts and their structural relationships can be used as important features for facial recognition, and these features are called geometric features. Knowledge based facial representation mainly includes methods based on geometric features and template matching.
Face Image Matching and Recognition
Face image matching and recognition: The extracted feature data of the face image is searched and matched with the feature templates stored in the database. By setting a threshold, when the similarity exceeds this threshold, the matching results are output. Face recognition is the process of comparing the facial features to be recognized with the obtained facial feature templates, and determining the identity information of the face based on the degree of similarity. This process can be divided into two categories: one is confirmation, which is a one-to-one image comparison process, and the other is recognition, which is a one-to-many image matching and comparison process.
