# Publication Citation: The use of slow motion' L\'{e}vy stable fractional diffusion smoothing in alternative methods of latent fingerprint enhancement}

NIST Authors in Bold

Author(s): Alfred S. Carasso; The use of slow motion' L\'{e}vy stable fractional diffusion smoothing in alternative methods of latent fingerprint enhancement} April 30, 2013 Photoshop processing\footnote{Mention of commmercial products or services in this report does not imply NIST appproval or endorsement of these products or services, nor does it imply that such products or services are necessarily the best available for the intended purpose.} of latent fingerprints is the preferred methodology among law enforcement forensic experts, but that appproach is not fully reproducible and may lead to questionable enhancements. Alternative, independent, fully reproducible enhancements, using IDL Histogram Equalization and IDL Adaptive Histogram Equalization, can produce better defined ridge structures, along with considerable background information. Applying a systematic {\em slow motion} smoothing procedure to such IDL enhancements, based on the rapid FFT solution of a L\'{e}vy stable fractional diffusion equation, can attenuate background detail yet preserve ridge information. The resulting smoothed latent print enhancements are comparable to, but distinct from, AFIS-ready Photoshop images. In addition, this progressive smoothing procedure can be {\em reexamined} by displaying the suite of progressively smoother IDL images. That suite can be stored, providing an audit trail that allows monitoring for possible loss of useful information, in transit to the user-selected optimal image. Such independent and fully reproducible enhancements provide a valuable frame of reference that may be helpful in informing, complementing, and possibly validating the forensic Photoshop methodology. NIST Interagency/Internal Report (NISTIR) - 7932 Public Safety/Security, Human Identification/Molecular Biometrics http://dx.doi.org/10.6028/NIST.IR.7932 Click here to retrieve PDF version of paper (2MB)