Palm Vein Pattern Recognition System

Palm Vein Pattern Recognition System

Author Name : V. Hema Lata, V. Divya Sri, M. Deepika, G. Roopa Krishna Chandra

ABSTRACT

The project palm vein pattern recognition uses hybrid texture descriptors such as discriminative robust local ternary pattern and Weber’s local descriptor for better recognition accuracy. Biometrics is substantially a pattern recognition system with biometric traits to recognize particular. The recognition performance on fingerprint and palm print biometrics has a drawback due tosome conditions such as oil on the fingers, moisture, and dirt. Therefore, vein patterns attract attention from the host of intrinsic biometric traits in developing a recognition system that can meet all the expectations. Vein patterns refers to pattern of blood vessels that appears as blue vessels located immediately underneath the skin that are almost invisible to the naked eye under natural lighting conditions and can be acquired in vivo only when employing infrared illumination, which effectively protects against possible external damage, spoof attacks, and impersonation. The texture of the blood vessels of different individuals are proven to be distinctive even among identical twins. The palm vein images are preprocessed, by selecting the region of interest for vein pattern extraction initially. Here, local thresholding is used to extract the vein pattern for its texture analysis by using two texture descriptors called Weber’s local descriptors and DRLTP are suggested to extract the features about texture to confess original templates. DRLTPprovides the shape and contrast invariant features of an object. WLD gives details about illumination changes between the pixels. Euclidean distance is used to match the features of test and original templates for a decision on person biometric. Finally the performance of the proposed algorithm will be measured with recognition accuracy and it provides a better matching rate than prior approaches.

Keywords: Discriminative Robust Local Ternary Pattern, Weber's Local Descriptors, Vein Pattern Extraction, Recognition Accuracy, Texture Analysis.