Application of a face recognition system based on LBP on android OS

Abstract

This paper presents the application of the algorithm of Local Binary Pattern (LBP) for face feature extraction and its implementation on Android operating system. The algorithm LBP is used for texture characterization and based on good performance was used to face characterization showing a good effectiveness, due to this is chosen to apply in a mobile device. To perform system testing on a mobile device was used a standard database (AR Facedatabase) to simulate the capture of images, which has 120 people and 21 images each, the average of 3 images was used for obtaining a template by person and using Euclidean distance it was obtained a 70% correct classification, showing that the LBP obtains good results using a simple classification algorithm with a limited processing power as have a mobile device, further tests were performed with 10 people where up to 94% recognition was obtained.