ABSTRACT The purpose of this project was to investigate the recent research work on, it was aimed to select a recent state-of-the-art bimodal approach for implementation, evaluation and possible improvement. Several state-of-the-art hand-based personal identification methods published during the last decade were reviewed. As a consequence of this review, the geometric features commonly used by these methods were identified hand-based personal identification; the focus was on those works adopting bimodal approaches that combine both hand geometry and palmprint features. It was also aimed to set up a proper hardware environment for acquiring hand images and building a database to be used in this research work as well as in any future work in the field of hand biometrics. Finallyand the major techniques adopted for describing palm-print features were examined. The review allowed us also to identify the main modules of a typical bimodal hand based personal identification system. A recent state-of-the-art hand-based personal identification approach using hand geometry and palm-print features proposed by Prasad et al (2009) was selected for implementation and evaluation and a new list of geometrical features was proposed to improve the performance of the selected method. In this approach, the geometrical features consist of 30 geometrical measurements of the hand and the palm-print features consist of the energies of the sub-bands resulting from a wavelet decomposition of the palm-print region. The matching process adopted in the approach is based on the nearest neighbor classifier. A database of hand images was built using our own acquisition environment. This database together with the Bosphorus database provided by Bogazici University, were used to test the performance of the algorithms implemented. It was found that on these two databases, algorithms based on geometrical features perform better than those based on palmprint features and as expected, bimodal methods outperform methods using single biometric. The results of our experiments showed that the new list of geometrical features improves the accuracy of the bimodal hand-based personal identification system.