Feature Extraction Model in Systems of Face Images for Person Identification
Abstract
One of the currently important and rapidly developing areas of modern information technology is related to the tasks of person identification by biometric characteristics (features). The advantages of face recognition systems include discreetness, passiveness, and relatively low cost. Despite all this, the issues of extracting the features of face images have not been adequately investigated. The paper considers the stage of face feature extraction, since it is the main stage in creating systems for person identification based on the features. To solve the problem under consideration, a model for extraction of the identification features of face images has been proposed. The basic idea of the proposed model is to form a set of preferred features that would simplify the process of building a decision rule when recognizing patterns given as images. To test the working efficiency of the proposed model, experimental investigations have been carried out in solving a model problem of face recognition.