A new hybrid approch for palm print recognition in PCA based palm print recognition system

The major issues that involve in identifying palm print are the search for the templates in the palm print database that best matches with the test sample from input. Here the fundamental to be solved are to select similar features of palm that are needs to be matched. The different feature of palm print that are able to discriminate them from each other must show a huge divergence between different users samples and few divergence between same user samples. For verification process principal lines and datum points are successfully used as an important palm print features, but some other features that are associated with a palm print are delta point features, geometry features, minutiae features and wrinkle features. By using the existing techniques, we propose a distinct scheme to facilitate the dynamic selection of palm print pattern to match it by combining various global and local features of a palm print in a hierarchical way. Our palm print matching system will operate in two steps, enrollment of user and verification of user. In first step that is enrollment; several palm print samples are obtained by a user to store them as templates in the system. Palm print scanner is used to capture the samples which then pass through preprocessing and feature extraction to create the templates which then stored in predefined palm print database. In second and final step, the palm print scanner is used to capture the fresh palm print sample of user. Then the captured palm print sample again passes through preprocessing and feature extraction. These extracted features of a user are compared with existing templates in the database to verify the identity of the user. In this paper, we are proposing a hybrid approach for palm print recognition using a combination of three different approaches of Image processing.