Font Size: a A A

An Iris-Based Personal Identification System

Posted on:2011-05-05Degree:MasterType:Thesis
Country:ChinaCandidate:H ZhengFull Text:PDF
GTID:2178360308462141Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
In the modern information society, higher requirements are needed in personal identification including accuracy, security and practicality. Traditional security technologies, such as tokens or password, have faced great challenges. Biometrics is about measuring unique personal features, such as a subject's voice, fingerprint, or iris. It has the potential to identify individuals with a high degree of assurance, thus providing a foundation for trust.. Iris recognition system is one of the biometric technologies, which is based on the features of our iris texture. An iris image is typically captured using a non-contact imaging device, which is of great importance and convenience in practical applications. Besides, iris pattern is unique which gives higher stability and security. Iris-based personal identification system is more accurate compared with other biometric technologies, such as face and fingerprint.An iris recognition system based on static images is designed in this thesis, including the iris location algorithm, iris normalization module, coding and matching algorithms based on Gabor filters. Furthermore, after introducing the iris image quality validation module, the video iris sequences-based personal identification system is accomplished.The main contributions of this thesis are as follows:(1) We modified the traditional iris location algorithm. Based on Daugman's integro-differential operator, by assuming the circular edge of the iris with a local convergence and using step-search method, modified algorithm greatly reduces the time needed to locate the iris for iris-based recognition systems and can be used in real-time applications. (2) A parameters estimation method was proposed for Gabor filters. For the normalized image size and filter size, the scope of filter parameters are limited, which makes it to set filter parameters automatically according to the used database, and increases the whole efficiency and convenience of the iris-based identification system.(3) We propose and implement a wavelet packet-based iris image quality evaluation method to extract effective iris image from the video iris sequences.(4) The iris-based personal identification system was built including the static images and video iris sequences. The system performances are analyzed.
Keywords/Search Tags:Iris recognition, Iris location, Image quality evaluation, Daugman's iris normalization, Integro-differential operator
PDF Full Text Request
Related items