Fingerprint quality assessment using frequency and orientation subbands of block-based fourier transform

A new fingerprint quality measure based on subband analysis of spatial-frequency domain is presented. The overlapped block-based Fourier transform is projected onto a new subband platform, which is related to frequency and orientations of ridges. Based on this platform, three proposed intrinsic spectral characteristics for fingerprint quality assessment can be extracted, i.e. global spectrum, relative spectral density, and relative spectral orientation continuity. The proposed method has an advantage of evaluating ridge spectrum directly, without suffering from incorrect parameter estimation such as directional field. The experimental results show that the discrimination capability of the proposed method is superior to conventional methods, especially those that employ thermal sensors.