Abstract

This paper presents a new and robust biometric authentication model utilizing palm print identification. The model incorporates advanced techniques including morphological Region of Interest (ROI) extraction with distance transform and un-decimated biorthogonal wavelet transform to ensure high levels of security. By leveraging the multi scaling capabilities of the wavelet transform, two distinct wavelet filter banks are employed to extract features from the distance transformed image. These extracted features are then compared with a test feature vector to determine the most effective feature factor for authentication purposes. Experimental results demonstrate the effectiveness of the proposed model, as it achieves a remarkable accuracy rate of 100% when tested with various images from the database. The combination of morphological ROI extraction, distance transform, and biorthogonal wavelet transform proves to be a powerful approach for palm print identification and enhances the security of the biometric authentication system. The findings of this study contribute to the advancement of biometric technologies and offer promising potential for real-world applications requiring reliable and highly secured authentication systems.

Description