U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Development of Pattern Recognition Methods for the Text-Independent Analysis of Handwriting Images

NCJ Number
86911
Journal
Schriftenreihe der Polizei-Fuehrungsakademie Issue: 1 Dated: (1982) Pages: 12-31
Author(s)
K Steinke
Date Published
1982
Length
20 pages
Annotation
This article provides a description of a computer-based method for the identification of handwriting specimens.
Abstract
The pattern analysis is based on such characteristics as the degree of slant and pressure, relative letter size, breaks between words and lines, the shapes of letters and their relationship to each other. The major contribution of the proposed method is the development of appropriate transformations of the handwriting image to a decision matrix. The method was used in an experiment that analyzed 4,000 handwriting specimens of 100 different writers. The recognition rate was 99.05 percent, which allows optimistic expectations for a successful practical application of the system. Charts, illustrations, and handwriting samples are given. The bibliography contains four entries.