Feature Extraction of Finger-vein Image Based on Morphologic Algorithm

As a new kind of identity authentication technology, finger-vein recognition has many merits. In order to improve the robustness and stability of detecting the finger-vein, an enhanced method for extracting finger-vein feature based on morphology is proposed in this paper, which combines with morphological peak and valley detection. The images will be scanned across the boundaries and the valley detection will be done from four directions. Such feature extraction method based on morphology improves the conventional point-wise comparison procedure and speeds up the operation. In the experiment, the finger-vein extracted by the proposed method has more precise details and better continuity comparatively. And with the noise added, the result of the proposed method makes the average error decreased. The experimental results show that the proposed method is effective and robust for the personal identification.