Warwick Warp » Technology
Low quality prints cause high failure rates
The success of an Automated Fingerprint Identification System (AFIS) depends on the quality of the fingerprints that are presented for enrollment and matching. Poor quality, low-contrast images and elastic deformation of fingerprints during acquisition are principal causes of failure-to-enrol (FTE) and failure-to-match (FTM). Minimising FTE and FTM rates is critical to any AFIS in order to reduce throughput time and associated costs.
Conventional image processing techniques are commonly used in AFIS to enhance images and improve the chance of matching, but they often struggle to deal effectively with the lowest quality and most distorted prints. Warwick Warp has taken a new, very different approach to the challenges that these issues present and has developed a patented preprocessing technology that significantly reduces FTE and FTM rates.
Fingerprint modelling - innovation for fast, accurate identification
The Warwick Warp approach centres on an innovative modelling technique which is unique in fingerprint identification. When presented with a live-scan or latent fingerprint, the preprocessor identifies and recovers ridge flow information and regularises both the flow and direction of the ridges. It removes artefacts and semi-permanent features such as scars or scratches and it reduces linear and nonlinear deformation, intrinsically aligning the image to produce a normalized, idealised model of the fingerprint with all its unique, identifying features preserved. This leading-edge modelling technique is the basis of the remarkable results that Warwick Warp is able to achieve with even very poor quality prints. Because the unique modelling process removes artefacts and aligns images at enrolment and not during matching, it increases matching performance in terms of speed and accuracy as well as significantly reducing failure-to-match rates and the costs associated therewith.
The technology works by applying a mathematical model to the print which identifies features of interest, such as ridges, and features not of interest, such as scratches or scars. The idealised model captures all the salient foreground information: contrast, singularities, ridge flow, minutiae and, if required, sweat pores, in a normalised coordinate system. It rejects the unwanted information and at the same time reduces or removes distortions, by understanding and recognising the characteristics of the ridges. In effect, it recreates the print as an idealised model that essentially matches the biological reality of the finger from which the print was obtained.
Please contact us to receive a white paper with more information about Warwick Warp's advantage.