Abstract: Nonlinear distortion is common in sensor-acquired fingerprint images. It degrades the performance of fingerprint verification systems severely. Especially for minutiae-based matching algorithm, which is most widely adopted, both the location and direction of minutiae are changed because of distortion. Based on the result of careful observation and examination on kinds of distorted fingerprint images，some useful characteristics of minutiae deviation on nonlinear distorted fingerprint image are found, and further used for minutiae adjustment. In this paper, a method to adjust the minutiae position and direction on the basis of multiple reference minutiae alignment method is proposed. However, other effective alignment methods can also be taken instead. Query fingerprint image and template fingerprint image are aligned first using multiple reference minutiae alignment method, and some landmark minutiae pairs from two images are computed. Then minutiae around each paired landmark minutia are adjusted, and comparison between adjusted query minutiae set and template minutiae set is made during the matching process. Compared with other typical methods, this method is easier to be added to existing minutiae-based matching algorithms while no additional feature but minutia is needed. Experimental results on FVC2004 DB2 show that this method can improve the matching performance clearly.