Automatic Face Recognition: Fuzzy Classification Versus Neural Networks

Abstract

In this paper, two automated face-recognition models are compared. The first model is based on feature extraction using histograms and elastic template matching. The extracted features are used to calculate a feature vector for each face. Feature vectors are passed to a fuzzy classifier which applies two distance measures for recognition. The second model employs more image-processing techniques in a complete vision system that is originally dedicated for three-dimensional object recognition. It mimics the human vision system by having hierarchical vision levels. In the early vision stage, it detects edges and extracts vertices. In the intermediate vision stage, perceptual grouping is performed and an invariant representation of the acquired image is passed to the next vision stage. The HAVNET neural network is employed in the third vision stage for learning and recognition. The performances of the two models are compared on a common database.

Department(s)

Engineering Management and Systems Engineering

International Standard Serial Number (ISSN)

0820-0750

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2024 The Authors, All rights reserved.

Publication Date

01 Dec 1996

This document is currently not available here.

Share

 
COinS