IDENTITY RECOGNITION OPTIMIZATION BASED ON LBP FEATURE EXTRACTION

ABSTRACT Unimodal systems have limited information that can be used for identity recognition systems. The multimodal system was created to improve the unimodal system. in this study is combination of face and palms at matching score level. Matching scores done using Weighted Sum Rule method. Extract features from each sample Local Binary Pattern (LBP) Meanwhile, large data dimensions are reduced by Principal Component Analysis (PCA) distance between palm measured closest distance, namely Euclidean Distance Benchmark dataset ORL, FERET PolyU. Based on testing database, an accuracy rate 98% (ORL PolyU) 95% (FERET obtained. test results show Hybrid method (PCA LBP) biometric runs well optimally. Keywords: Artificial intelegency, recognition, LBP,