Font Size: a A A

The Research And Realization Of The Face And Iris Biometric Identification Authentication System

Posted on:2012-01-13Degree:MasterType:Thesis
Country:ChinaCandidate:Y S LiuFull Text:PDF
GTID:2218330368487129Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Multimodal (such as face and iris, etc.) biometric identification technology is the field of pattern recognition in recent years, a very hot topic, which people own a variety of comprehensive utilization of biometrics, for today and provide more secure authentication methods . Researchers have been more and more attention.In this paper, the existing single-mode biometric system deficiencies, face and iris recognition on the composition of the multi-modal biometrics to do in-depth study, the main research work is as follows:1,The PCA-based and Adaboost face recognition algorithm were in-depth study;2, The Adaboost and SVM-based iris recognition algorithm were in-depth study;3, The various face-iris recognition fusion algorithms models were more in-depth study; According to actual needs, propose a new hybrid fusion model. Feature level fusion based on the actual algorithm is given.4, designed based on the DM642 + ARM2440, dual-core structure, embedded, face - recognition, integration, and multi-functional, the prototype system. Gives, in the development process, the resulting system, the needs analysis; hardware platform design, circuit and circuit board design; and software design flow, and UML modeling diagram.Also demonstrated in the development process, designed to verify the algorithm, PC-based human face - iris recognition algorithm verification experiment system.
Keywords/Search Tags:Multi-modal biometrics, Face recognition, Iris recognition, Feature fusion, Support vector machine SVM, AdaBoost, Embedded Systems
PDF Full Text Request
Related items