Quantification of the shape of handwritten characters: a step to objective discrimination between writers based on the study of the capital character O.

In view of contributing to the scientific validation of the individuality of handwriting, the testing of the two so called fundamental laws of handwriting--1: no two people write exactly alike; 2: no one person writes the same word exactly the same way twice--was approached by analysing the shape of 445 handwritten capital characters O produced by three individuals. A methodology based on classical Fourier descriptors was applied to the characters contours, which were extracted through an automated procedure of image analysis. Precise individual characterization of the shape was possible through Fourier analysis. Within-writer variability of the shape of character O for the writers selected could be shown in an objective and quantitative way through the statistical analysis of the Fourier descriptors. It was demonstrated that this polymorphism differed between the three writers. Differentiation between writers was quantitatively demonstrated by discriminant analysis of the Fourier descriptors, and by the existence of marked morphological distances between the set of characters O of each writer. The degree of dissimilitude of the character O writings could, thus, be assessed. Because of relatively reduced within-writer variability and a pronounced differentiation between the writers, a morphological profile could be established and discrimination between writers could be obtained through the quantification of the shape of one handwritten character.

[1]  Mark Denbeaux,et al.  Exorcism of Ignorance as a Proxy for Rational Knowledge: The Lessons of Handwriting Identification Expertise , 1989 .

[2]  Wilson R. Harrison,et al.  Suspect Documents: Their Scientific Examination , 1958 .

[3]  G. G. Stokes "J." , 1890, The New Yale Book of Quotations.

[4]  R. Muehlberger,et al.  A Statistical Examination of Selected Handwriting Characteristics , 1977 .

[5]  Linda R. Taylor,et al.  A System for Handwriting Classification , 1987 .

[6]  Pete E. Lestrel,et al.  Fourier Descriptors and their Applications in Biology , 2008 .

[7]  A. M. Wing,et al.  The variability of cursive handwriting measure defined along a continuum: letter specificity , 1987 .

[8]  Robert J. Muehlberger,et al.  Survey of Handwriting Habit Areas Used by Forensic Document Examiners: Degree of Use and Discriminatory Power , 1999 .

[9]  W. Kuckuck Computer-Aided Identification of Handwriting , 1991 .

[10]  R. A. Hardcastle,et al.  A computer-based system for the classification of handwriting on cheques. Part 2: Cursive handwriting , 1990 .

[11]  Sung-Hyuk Cha,et al.  Individuality of handwriting. , 2002, Journal of forensic sciences.

[12]  Tieniu Tan,et al.  Personal identification based on handwriting , 2000, Pattern Recognit..

[13]  J. Santamaría,et al.  Comparative Morphometric Study of a Human Phalanx from the Lower Pleistocene Site at Cueva Victoria (Murcia, Spain), by means of Fourier Analysis, Shape Coordinates of Landmarks, Principal and Relative Warps , 1996 .

[14]  R. N. Totty,et al.  A Computer-based System for the Classification of Handwriting on Cheques , 1986 .

[15]  Guojun Lu,et al.  Shape-based image retrieval using generic Fourier descriptor , 2002, Signal Process. Image Commun..

[16]  M. Schmittbuhl,et al.  Shape of the piriform aperture in Gorilla gorilla, Pan troglodytes, and modern Homo sapiens: characterization and polymorphism analysis. , 1998, American journal of physical anthropology.