IDENTIFIKASI CITRA TANDA TANGAN MENGGUNAKAN LOCAL BINARY PATTERN (LBP) PADA SMARTPHONE BERBASIS ANDROID

Abstract

Signatures are currently used by the majority of the public to authorize a document or as proof of transaction. The signature is a unique writing style of someone's name or as an identification sign of that person. This application uses the Local Binary Pattern (LBP) method as its feature extraction and Manhattan Distance as recognition. The LBP operator is a texture descriptor that uses a comparison of gray values of neighboring pixels. For training, data retrieval is done by collecting data 10 times from 50 participants with the amount of training data of 500 samples. The test is done 5 times each participant with the test results on average by 80.8% and the highest test by 94%.