General purpose bivariate quality-metrics for fingerprint-image assessment revisited

We evaluate a set of general purpose bivariate image quality measures with respect to their ability to assess fingerprint-image quality. Our evaluation approach is able to identify specific weaknesses or strengths of quality indices with respect to certain types of quality impairments. To systematically generate corresponding test data, we apply StirMark image manipulations (as being developed in the context of watermarking robustness assessment) which are able to simulate a wide class of acquisition conditions and qualities, applicable to any given dataset and also of potential interest in forensic analysis. Experimental results document the competitiveness of the general purpose image quality measures in a comparison with dedicated univariate fingerprint-image quality measures.