Singular Value Decomposition (SVD) Berdasarkan Intensitas Pencahayaannya Untuk Pengenal Wajah

Abstract

Singular Value Decomposition (SVD) is a method that can be used for biometric systems, where the biometric system is a system for identification by using physical features or human limbs such as fingerprints, eye retina, face, and others. This writing aims to recognize faces based on the intensity of their lighting using SVD. The use of SVD is done by forming the basis of a matrix (where this matrix is a collection of database faces (db)), then this base is used to transform the inputted image / face / db file and database files, then the results of this transformation the norm is calculated, if the minimum norm (db files and input files) is still within a certain tolerance the input file will be categorized as face / db file / ith person. The simulation results show that SVD can classify an image as a face or not with an accuracy of 94.7231% and a period of 35.2020 seconds for the recognition of 900 image files, a face image is a database or not of 68.4616% within 7.5480 seconds for the recognition of 650 face files, a The face images that are in the database are 90.7692% of the people in the 3,7340 seconds period for the introduction of 325 database files.