Image analysis – Pattern recognition – On-line recognition of handwritten characters
Reexamination Certificate
1999-06-22
2002-09-03
Boudreau, Leo (Department: 2621)
Image analysis
Pattern recognition
On-line recognition of handwritten characters
C382S115000, C382S116000, C382S119000, C382S181000, C382S186000, C382S188000, C382S154000, C382S202000, C382S218000, C382S313000, C382S314000, C902S003000, C902S005000
Reexamination Certificate
active
06445820
ABSTRACT:
BACKGROUND
a. Field of the Invention
The present invention relates generally to methods for the analysis of a person's handwriting, and, more particularly, to analysis of handwriting using a computer-assisted methodology.
b. Background Art
The analysis of handwriting, referred to as graphoanalysis, is ordinarily conducted for one of two purposes: (a) comparison of handwriting samples to determine whether or not the writer is the same person in both instances, and (b) evaluation of the person's writing relative to predetermined criteria to ascertain one or more aspects of the writer's personality, such as emotional characteristics. Examples of the first category often include law enforcement work, such as the determination of whether or not a signature or document is a forgery, or determination of whether a document was written by a suspected person. Examples of the second category, in turn, include evaluation of a person's personality and emotional responsiveness to determine their suitability for employment in a position requiring particular skills or personality traits, or their assignment to work with a certain group of people or to perform certain tasks.
Both categories of analysis require obtaining extensive, painstaking measurements on one or more handwriting specimens. In the first category, that of determining whether or not a particular person wrote a certain document, numerous, often minute details of the person's writing must be measured and cataloged, much in the manner of fingerprint analysis. In the second type of work, that of determining a person's personality or emotional characteristics, multiple measurements of certain features of the writing are taken and then analyzed statistically for comparison with the predetermined standards, which for the most part have been derived on an empirical basis from the analysis of handwriting produced by various persons having known personalities or emotional characteristics.
While graphoanalysis for these and other purposes is a valued tool in many industries/fields, its widespread use has become increasingly hampered by the fact that the necessary measurements have up to this time been performed almost invariably by manual means, using a magnifying glass, ruler, protractor, pencil, and other unsophisticated tools. For example, to perform a personality analysis, the graphoanalyst must use a ruler or similar guide to determine the base line of the writing, visually determine certain key points on various letters, and then measure the slant angle of scores or even hundreds of these letters visually using a protractor; as with most forms of statistical analysis, a fairly high number of measurements is required to form a basis for a reliable analysis. The great deal of time and labor which must thus be spent to professionally analyze even a single person's handwriting, especially in view of rapidly rising labor costs, has often rendered this otherwise valuable tool uneconomical for use in all but the most important cases. Moreover, since the manual measurement techniques require drawing various lines and marks on the writing sample using a pencil or other writing instrument, these necessarily deface and/or damage the original to one extent or another, which renders obtaining subsequent measurements (e.g., by a second graphoanalyst) more difficult and otherwise decreases the usefulness of the original document.
Perhaps an even more serious problem is the degree of variability and sometimes inaccuracy which is inherent in the conventional, manual graphoanalysis techniques. Human judgment and therefore human error is invariably involved to some extent in such techniques, and therefore the quality of the analysis is heavily dependent on the manual skills of the individual graphoanalyst. Moreover, since each analysis often requires hundreds of measurements, fatigue often become a significant factor, and can impair the efforts of even the most skilled practitioner. Still further, determining the base line and measurement points on the handwriting specimens is a high subjective process, which results in a high degree of variability between the measurements taken from the same sample by different analysts. Not only do these various factors impact the accuracy of each analysis, they also make it difficult to properly compare the measurements to the precise standards which are necessary for a proper determination of personality/emotional characteristics.
As a result, although the value of graphoanalysis is well established, particularly in Europe (for example, handwriting analysis is used in employment screening for 40% of job applicants in Great Britain, and for 80% of applicants in France and Israel), the inefficiencies, inaccuracies and variabilities which are inherent in the manual measurement techniques have stymied its further, widespread application. For example, graphoanalysis is potentially an extremely valuable tool for the human resources departments of commercial enterprises and governmental agencies, to help determine the suitability of a person for employment or assignment to a particular position or team, but the existing problems with cost and accuracy have thus far limited its adoption in these arenas.
Similarly, the difficulty in obtaining economical, accurate analysis of handwriting specimens has rendered this resource unavailable to many criminal and civil investigators, and this has been a particular problem for police departments which are located outside of major metropolitan areas, where both the availability of skilled graphoanalysts and departmental budgets are often limited.
Accordingly, there exists a need for a method for measurement one or more characteristics of a handwriting specimen which does not require these measurements to be performed manually, and which therefore eliminates the element of inaccuracy and variability in these measurements. Furthermore, there exists a need for such a method which enables large numbers of such measurements to be taken, compiled, and analyzed quickly and economically. Still further, there exists a need for such a method which enables such measurements to be taken in a standardized manner, so that these can be compared with confidence to precise, predetermined standards which assign personality characteristics or other elements to such measurements. Still further, there exists a need for such a method which permits such measurements to be taken and used by a trained graphoanalyst who is not necessarily located in the vicinity as the client or other requester, so as to make this resource more readily available to entities located outside of major metropolitan areas.
SUMMARY OF THE INVENTION
The present invention has solved the problems cited above, and is a method for accurate and quantitative analysis of a handwriting sample.
The first step in the method is to create a digital bit-map of the handwriting sample, as by using an electronic scanner or digital camera. The bit-map file is then used to create a digital image of the writing sample, and a cursor is used to mark selected points on elements of the writing for measurement. The measurements include determination of the slant angle of strokes in the handwriting and measurement of heights of the major areas of the writing.
The measurements are tabulated and/or categorized according to a predetermined scheme, and these results are then compared with a predetermined standard for determining certain characteristics relating to the person who produced the handwriting sample.
The slant angle measurements can be made by using the cursor establish a base line for each stroke and a second line between a starting point where the stroke lifts off the base line of the stroke, and an ending point where the stroke ceases to rise, and then calculating the angle between the two lines. The cursor may also be used to mark the tops of the letters for measuring the height of the areas of the writing, by calculating the vertical distance between the tops of the letters and base line.
The moving cursor m
Boudreau Leo
Limbic Systems, Inc.
Mariam Daniel G.
Schacht Michael R.
LandOfFree
Method for conducting analysis of handwriting does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for conducting analysis of handwriting, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for conducting analysis of handwriting will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2894292