Fingerprints is one of biometric identification system. This is because fingerprints have unique and different pattern in every human, so identification using fingerprints can no longer be doubted. But, manual fingerprint recognition by human hard to apply because of the complex pattern on it. Therefore, an accurate fingerprint matching system is needed. There are 3 steps needed for fingerprint recognition system, namely image enhancement, feature extraction, and matching. In this study, crossing number method is used as a minutiae extraction process and template matching is used for matching. We also add alignment point pattern process added, which are ridge translation and rotation to increase system performance. The system provide a performance of 18,54% with a matching process without alignment point pattern, and give performance of 67,40% by adding alignment point pattern process.