In this paper, a new fingerprint verification algorithm is presented that improves matching accuracy by overcoming the shortcomings of previous methods due to poor image quality. It reduces multi-spectral noise by enhancing a fingerprint image to accurately and reliably determine a reference point using the orientation reliability and then extract a 129 × 129 block, making the reference point its center. From the 16 co-occurrence matrices, four statistical descriptors are computed. The experimental results have been analyzed using FVC testing protocol; the equal error rate (EER) is 0.32%. Furthermore, the comparison with other methods shows that the proposed method is more accurate and robust for reliable fingerprint verification.