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 are divided 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.