Font Size: a A A

Research On Subspace Methods For Palmprint Recognition

Posted on:2011-10-04Degree:MasterType:Thesis
Country:ChinaCandidate:D F ZhuFull Text:PDF
GTID:2178360332955716Subject:Applied Mathematics
Abstract/Summary:PDF Full Text Request
With the rapid development of information technology and network technology, there is an unprecedented importance that effective identification of the personal true identity. Biometrics has emerged to solve such problems, and is found to be one of the most important and effective solutions for this task. While palmprint recognition, developed in recent years, is a new biometrics technology. Palmprint recognition has attracted domestic and foreign researchers attention, because of rich information, difficult to counterfeit, high acceptance, the small noise, high accurate recognition rate. The start of palmprint recognition late, and the investigation of its theory and application should be enhanced and improved further. Thus, it is very necessary and impending for ones to make a deep study on palmprint recognition.This paper studied a large number of domestic and international academic literature on the palmprint recognition, and depth study the subspace-based feature extraction method for palmprint recognition. Experiments show that the improved feature extraction algorithm for palmprint recognition is effective and feasible. The main work in this thesis can be summarized as follows:First, this paper introduced the common classical subspace method, and compared the per-formance of based on subspace method for palmprint recognition, then compared and analysed the classical subspace method.Second, based on the analysis of the Fourier and LDA, an effective algorithm is proposed for palmprint recognition combining the Fourier Transform and LDA. We exploit only the center bandwidth of the Fourier magnitude for recognition. The results of experiments conducted on PolyU palmprint database demonstrate the effectiveness of proposed method. All in all, the proposed method that is robust against illumination is a suitable even wonderful method for palmprint recognition.Third, on the basis of MFRAT and ICA, we propose to carry out palmprint recognition by combining MFRAT and ICA. MFRAT can effectively extract the palm lines from the complex palmprint images, then use the ICA algorithm to do feature extraction. The experimental results showed that MFRAT+ICA for palmprint recognition has greatly improved compared to ICA.
Keywords/Search Tags:Biometrics, Palmprint recognition, Subspace method, Directional representation, MFRAT
PDF Full Text Request
Related items