A Systematic Algorithm for Fingerprint Image Quality Assessment

A New Ridge-Features-Based Method for Fingerprint Image Quality Assessment

“…Some features used for quality assessment are based on pixel intensities like Local Clarity Score (LCS) [4], low contrast map [10], gray intensity mean, gray intensity standard deviation [2][8] [12], uniformity, smoothness, inhomogenity [2][12] and texture features [13]. Other features are extracted from the spectral domain like power spectrum and the response of Butterworth band-pass filters using Fast Fourier Transform [8], or global spectrum and relative spectral density [9].…”

“…On the other hand, features based on the orientation field are also used. Examples of this are orientation certainty level (OCL) [6], Local Orientation Quality Score (LOQS) [4], direction map, low flow map, high curve map [10], orientation coherence [12], relative spectral orientation continuity [9], orientation certainty and consistency [2] and ridge-line smoothness [11]. Also, penalty due to the backgrounds noise and the quality of the core point position are features used [12].…”

“…Some approaches compute a quality value for each analysed block [2][6] [8][10] and in a final step an overall quality value is obtained by combining the blocks classification results. In other cases a combination of the features values is used to estimate the final quality measure [12].…”

A Multiple Indexes Quality Assessment for Fingerprint

“…On the other aspect, estimation method was explored by using single index analysis, multivariate linear weighted [7] and multi-index nonlinear fusion [8]. Artificial Neural Network (ANN) and Back Propagation Neural Network (BPNN) based nonlinear classifiers were developed [9], respectively.…”