A Facial photos in today's era are widely used as a media for identity recognition, but not many computer applications provide identity recognition of face photos that contain the names of the photo owners, to make an a prototype the sistems use a KNN algorithm, this algorithm work is by classifying the closest object and grouping it on predetermined objects. In this paper, the object is a face photo where the KNN algorithm will be used to classify the facial patterns contained in the photo. The stages in pattern recognition, starting from preprocessing, feature extraction and then classification. In addition to using the KNN algorithm for data classification, photo of faces will be detected and stored the T-Zone area and frontal face. In this paper 11 images used for data testing and the accuracy will be calculated using a recognition algorithm. The results of this paper are a facial recognition program using python that can display faces with a validity level of 82%.