원문정보
초록
영어
Biometric has been implemented on numerous public facilities to enhance the security system. Fingerprint and face are the most popular biometric. Emerging technology has introduced potential biometric such as palm print, lips, teeth, vein and ear. However, most of this biometrics requires a special device to capture it. Thus, the implementation of such system will be costly. Iannarelli [1, 2] has proved that ear biometric is having a great potential for identifying a person. In this research work, an attempt is made to improve the detection and finally to segment the human ear from the whole image of human’s head. The success of this stage is very important for achieving the later goal, such as recognition and classification. This paper introduces a novel method for ear segmentation. Proposed method is based on morphological analysis fused with RBF neural network. Experiment shows that the proposed method has delivered a promising result.
목차
1. Introduction
2. Proposed Hybrid Segmentation Method
3. Development of a Pre-processing Method
4. Development of a Classifier for Ear Edge Detection (s)
5. Development of a Post-processing Method
6. Experimental Results and Analysis
7. Conclusions
References