Fingerprint technique is one of the most reliable biometric techniques. Reliable extraction of features from poor quality prints is the most challenging problem faced in the area of fingerprint recognition. In all of the previous fingerprint recognition methods, pre-processing such as smoothing, binarization, and thinning and other post processing operations are required These systems are not useful for the compressed or noisy images. Wavelet based algorithm is superior in terms of both recognition rate and the computational complexity, this algorithm is the key to making a low cost and fast fingerprint identification system that would operate on a small computer system. In this work, a fingerprint recognition system has been developed which is based on the texture pattern of the fingerprint. In the proposed approach, features are extracted directly from the original image without preprocessing, thus are computationally efficient. Implementation is done on MATLAB R2010b as the programming tool.