#TITLE_ALTERNATIVE#
<p align="justify">Face recognition is the currently the most developed biometric system. Biometric is a system which uses physiological and behavioural characteristics to recognize the identity or to verify the claimed identity of a person through automated means. Face recognition i...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/13467 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | <p align="justify">Face recognition is the currently the most developed biometric system. Biometric is a system which uses physiological and behavioural characteristics to recognize the identity or to verify the claimed identity of a person through automated means. Face recognition is developed because it has a high comfortability and acceptability rate. Face recognition is a part of pattern recognition which uses face image as input. Face recognition system is divided into four phases. The phases are image processing, face detection, feature extraction and classification.<p align="justify"><p>The focus of this paper is on the extraction and classification phase which extract the image features and classify it. The input is the image from face detection phase that only detects image that contains a face of a person. Extraction feature transforms the image feature to a feature that has smaller dimension (feature space) using Fisherface methods. Fisherface is a combination PCA (principal component analysis) and FLD (Fisher Linear Discriminant). The classifier which is used in this system is a multilayer neural network using backpropagation algorithm.<p align="justify"><p>Tests conducted in this system resulted better than previously developed Eigenface method. According to the simulation, the integrated system of both face detection and recognition has resulted in 83% accuracy rate. |
|---|