Browse Prior Art Database

Finger Print Identification

IP.com Disclosure Number: IPCOM000096551D
Original Publication Date: 1963-Jul-01
Included in the Prior Art Database: 2005-Mar-07
Document File: 2 page(s) / 42K

Publishing Venue

IBM

Related People

Hillsley, RH: AUTHOR

Abstract

The device effects two-dimensional matching. The two-dimensional data patterns are on record member 1 against which the unknown data on record member 2 is to be matched. The data on member 1 is preprocessed to derive a limited number of correlation constants which are stored in memory 3. Sample observations taken from the unknown analog data are converted by A to D converter 4 and are digitally processed by computer 5 with the correlation constants. If adequate correlation is detected, the resultant computation yields the XY displacement between the sample observations and the stored data. These XY displacement signals are converted by D to A converter 6 to cause movement of record member 2 for subsequent correlation testing.

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 81% of the total text.

Page 1 of 2

Finger Print Identification

The device effects two-dimensional matching. The two-dimensional data patterns are on record member 1 against which the unknown data on record member 2 is to be matched. The data on member 1 is preprocessed to derive a limited number of correlation constants which are stored in memory 3. Sample observations taken from the unknown analog data are converted by A to D converter 4 and are digitally processed by computer 5 with the correlation constants.

If adequate correlation is detected, the resultant computation yields the XY displacement between the sample observations and the stored data. These XY displacement signals are converted by D to A converter 6 to cause movement of record member 2 for subsequent correlation testing.

The technique employs the substitution of the complete multidimensional data by stored data in the form of correlation constants which yield the centroid of the data correlation function. The location of the centroid is the X, Y, Z - - displacement which is sought. The correlation constants are mathematically derived to minimize the data required to be stored, representing the known data, and to facilitate straight-forward digital operations so as to speed and simplify required calculations.

This technique has application to numerous matching operations such as registration of alpha-numeric characters, map-matching, accurate determination of sensor position, testing cams for machining accuracy, given Z (A, B, C--)...