ANALISIS DAN SIMULASI SISTEM PENGENALAN WAJAH DENGAN METODE FISHERFACE BERBASIS OUTDOORVIDEO

Nurani Fitriyah, Telkom University and Dr. Ir. Bambang Hidayat, Telkom University and SuciAulia, ST,MT, Telkom University (2015) ANALISIS DAN SIMULASI SISTEM PENGENALAN WAJAH DENGAN METODE FISHERFACE BERBASIS OUTDOORVIDEO. In: Seminar Nasional Universitas PGRI Yogyakarta 2015.

Preview

Text 2TI.12_Nurani Fitriyah 433-438.pdf Download (1MB) | Preview

Abstract

Facial recognition is currently the most developed biometric system that can identify a specific individual in a digital image by analyzing and comparing pattern. Conventional surveillance video recording does not have capability to recognize people automatically. The design of human face recognition as an added value of surveillance video recordings installed in an open space is an appropriate solution.In the previous experimental using Fisherface in image processing has resulted in 83% accuracy rate. Success of Fisherface methods is reached by combining FLD (Fisher Linear Discriminant) and PCA (Principal Component Analysis). In this paper, Fisherface is implemented for face recognition in outdoor video processing with the duration is more than 7 seconds and have 15 frames per second each video. The experiment used captured video sample frames as train database and video as test database. The classifier which is used in this system are Artificial Neural Network (ANN) using Backpropagation algorithm and also Euclidean Distances a comparison. This experimental result in video processing shows that face recognition method proposed in this paper combined with ANN classifier makes impressive performance improvement proved with high accuracy compared with conventional Eigenface methods.

Actions (login required)

View Item