IJASEIT

International Journal on Advanced Science, Engineering and Information Technology, Vol. 12 (2022) No. 3, pages: 1222-1228, DOI:10.18517/ijaseit.12.3.11666

Face Recognition Application Based on Convolutional Neural Network for Searching Someone’s Photo on External Storage

I Putu Arya Dharmaadi, Deden Witarsyah, I Putu Agung Bayupati, Gusti Made Arya Sasmita

Abstract

Digital photos are often defined as personal archives collected long ago and are stored on a large enough storage media such as an external hard disk or flash disk. Problems arise when someone wants to find photos of themselves or others in tons of photo collections. Searching manually, such as opening a photo file or folder one by one, will certainly be very troublesome. Based on these problems, this study designed an application for searching certain photos based on the similarity of the inserted face photo. This application is built for computer or laptop devices, which was developed by using the Python programming language and Dlib module that applied the face recognition method through the combination of Convolutional Neural Network (CNN), FaceNet Embedding, and Triplet Loss for matching faces. The recognition scheme starts from face detection, face alignment, face encoding, and face classification stage. Our application is very handy to run in looking for particular face images on external storage compared to prior studies. We have done experimental research, demonstrating that the application can find almost all image files the user is looking for. In addition to the result in the form of an application, this study contributes to exploring the performance of the Dlib module, in terms of precision and recall rate, which could not recognize non-frontal face images well. We encourage other researchers to address this limitation in further studies.