人耳人脸融合识别系统预处理模块设计.doc

人耳人脸融合识别系统预处理模块设计
院 系
自动化学院
专 业
自动化
班 级
学 号
姓 名
指导教师
负责教师
沈阳航空工业学院
2009年6月
摘 要
生物特征识别推向大规模实际应用的趋势和需求，使得多模态生物特征识别技术成为生物特征识别领域的研究和应用热点。基于多模态生物特征融合进行身份鉴别可以实现多种生物特征的优势互补，从而在提高准确率、扩大适用范围、增强系统可靠性等方面提高生物特征识别系统的性能，使之更接近实用。目前，人耳人脸融合识别已成为当前多模态识别的研究热点。在人耳人脸融合识别系统中，人耳人脸图像的预处理占有十分重要的地位，图像预处理的好坏直接影响人耳人脸融合识别系统识别的正确率。
本课题的主要内容是设计人耳人脸融合识别系统的预处理模块。人脸图像预处理模块采用基于代数特征的方法；人耳图像预处理模块也采用基于代数特征的方法，但需借助PCA的识别方法来验证人耳图像的预处理效果。
关键词：人耳人脸融合识别；生物特征识别；图像预处理；代数特征；几何特征；
Abstract
The trends and needs of biological identification to the large-scale practical application, make multi-modal biological identification technology as the research and application highlight in the field of biological identification. identification based on multi-modal biological feature integration can realize predominance complement in a variety of biological characteristics. And this technology can improve the accuracy and expand the scope of application, and enhance system reliability and the forth, which can improve the performance of the identification system and make it more practical. At present, integration of the human ear Face recognition has become hot spot in the study of multi-modal identification. the human ear face image preprocessing is of great important status in the field of the human ear face integration identification , image pre-processing has a direct impact on the accuracy of human ear face integration Identification System.
The main content of this issue is to design preprocessing module of the human ear face integration Recognition System. Face image preprocessing module is based on the algebra features, and the human ear image pre-processing module is based on the algebra features too, but which utilize PCA identification method to verify the results of human ear image preprocessing.
Keywords: integration of the human ear Face recognition; biological identification; image pre-processing; algebra features; geometrical features;