Local-global-based palmprint verification

2D palmprint based biometric system can effectively and automatically recognize, with a high confidence, a person's identity. In such a system, feature extraction and matching serves as a key component since it has a direct effect on the overall performance of the system. In this paper, we propose a unified framework to combine the local feature and the global feature to improve the system's performance. Specifically, the orientation information extracted by the Gabor filters is coded as the local feature. By increasing the scale of Gabor filters to infinite, actually we can get the Fourier transform of the image, and hence the Fourier transform coefficients of the image can be taken as the global features. Such kinds of local and global features are naturally linked via the framework of time-frequency analysis. The BLPOC function between two images is used as a measure for the global features. Experiments are performed on the PolyU palmprint database. The promising results demonstrate the effectiveness of the proposed scheme.