Hand geometry has been a widely used biometric authentication because it is generally believed that the human hand has sufficient anatomical features which could be used for personal identification. Many hand geometry systems use pegs, which guide hand placement on the scanner. The system prompts the user to position the hand on the scanner several times and only captures when the current position is satisfied. In such a system, measurements are not very precise and this reduces accuracy during feature extraction. The system also has a higher false acceptance rate. This paper presents a peg-free hand geometry recognition system that does not depend on the orientation of the hand. Several features from test hand images are extracted and stored in the database, which are used to train an artificial neural network (ANN). To facilitate easy usage of the hand geometry verification system (peg-free), a GUI was developed using MATLAB software. The developed system was validated and the overall result shows that the system can be used for biometric verification using hand geometry where the orientation and placement of the hand are not a necessity. The results show that the developed system performed better with a relatively low false acceptance rate and false rejection rate of 0.01% and 0.02% respectively. The system also has a lower mean square error of 8.84×10-5.