A new approach for fingerprint verification, based on wavelets and pseudo Zernike moments (PZMs), is discussed. PZMs are robust to noisy images, invariant to rotation and have a good image reconstruction capability [4]. PZMs have been used for global analysis and so they are used to extract global features (the shape of the fingerprint image). Wavelets are good at local analysis and so they help to extract local features (minutiae) from a fingerprint. Therefore, this hybrid approach extracts most significant features from the fingerprint images and achieve better verification rate. Different types of wavelets are used for the study but the result shows that Symmlet orthonormal wavelet of order 8 gives best verification rate.