| With the awareness of security issues enhanced, traditional identification technology can not meet current requirement. Biometrics technology can fit for the need of modern security system. Today, with the rapid development of information and internet all over the world, information security is extraordinarily important, especially, personal identification technology. Because biological feature identification technology has unique stability, uniqueness and convenience, it is applied more and more broadly. As palmprint is typical of ease acquisition, and its major feature is apparent, stability and divisibility, palmprint automatic identification is a very potential identification technology.Through the study and research to the development history and current research to this subject, we know that this technology is still in a stage of rapid development. On the base of the main technologies in this field, we mainly focused on the palmprint recognition technology on the base of frequency domain. This paper we proposed two new methods based on frequency domain. One is based on band-limited phase correlation technology, the other is based on 2-D log Gabor filtering with feature fusion. The first method improved the traditional phase correlation technology by eliminating the meaningless high frequency components. The experiment result shows that it is not only better than the phase correlation, but also get high recognition rate under lower resolutions. For the second method, Multi-orientational 2D log Gabor filters are employed to extract the palmprint features. To reduce redundancy of the log Gabor features, multiple log Gabor features with different orientations are combined by using the magnitude information, and the fusion features are encoded based on the phase information. The similarity of two palmprint codes is measured by their improved Hamming distance. The validity has also been demonstrated and the result shows that it is better than the traditional Gabor function. |
PDF Full Text Request