Article Per Year (5 Year)
p-Index From 2019 - 2024
0
P-Index
This Author published in this journals
All Journal SENTIA 2015
Mila Fauziyah
Seminar Nasional Teknologi Informasi Politeknik Neheri Malang
Education

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
PENENTUAN NILAI VEKTOR PEWAKIL AWAL PADA ARSITEKTUR JARINGAN SYARAF TIRUAN LVQ UNTUK PENGENALAN WAJAH Devira Anggi Maharani; Mila Fauziyah; Denda Dewatama
SENTIA 2015 Vol 7, No 2 (2015)
Publisher : SENTIA 2015

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (696.612 KB)

Abstract

Artificial Neural Network (ANN) has special ability to do recognition of large, dynamic, and non-linar system through learning technique which could not be done by face recognition system using mathematical formulation methods. In this application, face recognition technique needs number of dynamic dimensions for the determination of model in this system. Therefore, mathematical methods are often not effective to resolve these problems. In the development of technology about face recognition, artificial neural network continues to develop, particularly in terms of fast and reliable. One type of familiar and commonly applied artificial neural network is Learning Vector Quantization (LVQ). This research will use LVQ as method of face recognition because the training process is relatively faster than other methods of ANN. To improve the reliability of this method, the determination of initial weight in training process will be one of reference to provide a good level of accuracy and make this system convergent with faster time. System used in this research is a face image 100 x 100 pixel as matrix input, alpha values at interval of 0.1 in the 0.1-0.9 range, 1000 maximum epoch, 2 layers (input layer and output layer), initial weight data each class (10 classes), and LVQ can reach the highest recognition rate of 4 technique of weight value determination that achieves up to 100%.
Co-Authors Devira Anggi Maharani Dewatama, Denda