MULTIPLE FACE DETECTION AND RECOGNITION USING MACHINE LEARNING

In this background, we have used modes based on machine learning that allows a machine to develop through a learning process, and to accomplish tasks that are difficult or impossible to fill by more conventional algorithmic means. In this process, a face detection algorithm is used for extracting faces from photo frames (training photos) and to create a face database. In the next step, filtering and preprocessing are applied to face images acquired in the previous step. Finally, a collection of machine learning algorithms are trained using as input data the faces obtained in the previous step. Haar cascade classifier is used both for selection and for learning, which is a strong classifier with a cascade classification. Moreover, the classifiers are used for classifying faces obtained from video frames (test videos). The obtained result shows the suitability of this approach for analyzing large collections of videos where previous face labels are not available.