PENERAPAN BIOMETRIC FACE RECOGNITION MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK PADA APLIKASI BERBASIS ANDROID

Abstract

A security system with a high level of accuracy is needed to prevent data leakage, one way to prevent data leakage is to implement a face recognition biometric security system in a login system. In this study, the application of face recognition was carried out in the online workshop application login system (Mecha). The implementation is carried out using the Convolutional Neural Network method which is applied to android-based applications using tensorflow lite by adding assets with the .tflite extension to the Android Studio Integrated Development Environment (IDE) as well as the Google ML Kit library used as the pre-processing process. The results showed that the face recognition biometric security system can be applied to android-based applications and the Convolutional Neural Network method can be used to recognize faces even in dark conditions in online workshop applications (Mecha), the application of the Convolutional Neural Network method in online workshop applications (Mecha) provides average accuracy of 88.25%. With testing from genuine users of 99.25% and testing from non-genuine users of 77.25%.