Browse Prior Art Database

Fingerprint Image Enhancement System

IP.com Disclosure Number: IPCOM000080655D
Original Publication Date: 1974-Jan-01
Included in the Prior Art Database: 2005-Feb-27
Document File: 2 page(s) / 58K

Publishing Venue

IBM

Related People

Ting, YM: AUTHOR [+2]

Abstract

One of the most difficult tasks in fingerprint identification is to reproduce the print accurately in binary form. The TV scanner provides the analog video signal shown on Fig. 1. This signal rides on a DC noise signal which is due to the variations of the background density of the print. These variations might be attributed to a number of factors, such as smudging, variation in paper quality, over or underinking and paper soiling.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 70% of the total text.

Page 1 of 2

Fingerprint Image Enhancement System

One of the most difficult tasks in fingerprint identification is to reproduce the print accurately in binary form. The TV scanner provides the analog video signal shown on Fig. 1. This signal rides on a DC noise signal which is due to the variations of the background density of the print. These variations might be attributed to a number of factors, such as smudging, variation in paper quality, over or underinking and paper soiling.

The relatively rapid undulations which are superimposed on the slower background variation in Fig. 1, represent the true output signal of ridge structures of the print generated by the scanner. The local maxima correspond to ridges and the local minima to the valleys of the fingerprint. The background variations are too great for a fixed threshold level to define the ridges and valleys.

To overcome this problem, a dynamic threshold level as shown in Figs. 2a and 2b is established by means of the circuit in Fig. 3. The dynamic average level is established on a continuous basis, by averaging the values of vicinity points on the same line of the point of interest. For example, the average value at point i is va(i). The values of vicinity points are v(1 Tau), v(2 Tau), ---, v(nTau). Then:

(Image Omitted)

where B(i) is the binary representation of the signal at point i.

In Fig. 3, the values v(Tau), ..., v(nTau ) obtained by delay elements 10 are applied to the summing network comprising resistors R, R/n...