Font Size: a A A

Fingerprint Identification System Based On Multi-classifier And Pseudo-feature Tracking

Posted on:2007-03-09Degree:MasterType:Thesis
Country:ChinaCandidate:A ChenFull Text:PDF
GTID:2208360212479130Subject:Systems analysis and integration
Abstract/Summary:PDF Full Text Request
Accurate automatic personal identification is critical in a variety of applications in our electronically interconnected society. Biometrics, which refers to identification based on physical or behavioral characteristics, is being increasingly adopted to provide positive identification with a high degree of confidence. Among all the biometrictechniques, fingerprint-based authentication systems have received the most attention because of the long history of fingerprints and their extensive use in forensics.Due to the limited amount of information present in the minutiae-based representation, it is desirable to explore alternative representations of fingerprints. The research on fingerprint mainly based on two categories: (i) minutiae-based, and (ii) global-based. Because of there different theoretics and applications, normally they can't be realized into one system. Based on the research on classifier, we compared multi-classifier fingerprint system. To eliminate false minutiae point, we purposed an algorithm on tracing the linear missing of the image. This algorithm can be used to locate the scars and ruga. Based on the conception of the multi-classifier, we purpose a new method for fingerprint identification. The experiment shows that it is effective. By the mean time this algorithm could also to be used on locating the scars on the sensor.
Keywords/Search Tags:Biometrics, minutiae point, classifier, multi-classifier, false minutiae point, linear missing, linear missing tracing
PDF Full Text Request
Related items