Poor data quality is responsible for many or even most of the mismatched errors in fingerprint recognition systems. It became apparent that a particular effort is needed in adapting state-of-the-art fingerprint matching techniques to real-world conditions using quality measures. In this article we address a challenging problem of how to associate local quality measures with the descriptors of local minutiae, in particular Minutia Cylinder-Code (MCC), in order to obtain better recognition rates. First, we introduce a new local quality measure, called Cylinder Quality Measure (CQM), which corresponds to each MCC descriptor combining the qualities of the individual minutiae involved.