Image analysis – Pattern recognition – On-line recognition of handwritten characters
Reexamination Certificate
2000-02-28
2003-06-17
Chang, Jon (Department: 2623)
Image analysis
Pattern recognition
On-line recognition of handwritten characters
C382S198000
Reexamination Certificate
active
06580826
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to the processing of human handwriting for purposes of information storage, automatic character recognition, and the like. More particularly, this invention relates to the processing of spatiotemporally sampled symbols, including the representation of such symbols in parametric form.
ART BACKGROUND
The automatic analysis of human handwriting begins with sampling and digitization of the image or signal produced directly by human manipulation of a writing instrument, referred to herein, in a general sense, as a stylus. Purely graphical analysis can be performed on data sampled from a static image. However, for dynamic analysis, and more generally for analyzing data with reference to a temporal sequence, it is advantageous for the human subject to produce by manipulating an instrumented stylus or tablet that permits spatiotemporal sampling.
One exemplary instrumented tablet is described in U.S. Pat. No. 5,463,388, issued on Oct. 31, 1995 to R. A. Boie et al. This tablet includes a rectangular array of capacitance-sensing electrodes. The position of a hand-held stylus is determined, e.g., from the centroid of the respective capacitance values, as calculated in a microcontroller.
Parameter methods have been used for a number of years in connection, for example, with automatic signature verification. According to these methods, the signature (or other handwritten symbol) is represented in an abstract parameter space. The parametric representation consists of a set of numerical values of functions that are evaluated on the sampled data, and that relate to some combination of graphical and dynamic properties of the sampled data. Generally, a parametric representation is a condensed representation, in the sense that it occupies fewer bits of data-storage capacity than do the raw, sampled data.
Parametric representations of signatures have been used with some success for signature verification. In signature verification, the parameters are evaluated on a newly entered signature (or group of signatures), and the results are compared with a stored set of reference values. Such a procedure does not require the reconstruction, from parameters, of either the reference signature or the newly entered signature. Therefore, there is no need to choose parameters that preserve enough graphical information to reconstruct these signatures. Instead, parameters for signature verification are selected on the basis, e.g., of a tradeoff between selectivity and computational efficiency.
SUMMARY OF THE INVENTION
I have invented a parametric representation of handwritten symbols that not only permits efficient data storage, but also permits the symbols to be reconstructed, with a high degree of legibility, from the stored parameters.
Thus, in one broad aspect, my invention involves a method for representing a handwritten symbol in parametric form. This method comprises obtaining a temporally sequenced record of data points. Each of these data points describes the x and y coordinates of a sampled point on the handwritten symbol, and also includes a pen-condition flag that describes whether the pen was in the pen-up or pen-down position when that data point was recorded. (Instrumented tablets are available that will indicate not only whether the pen was up, but also what the pen pressure was on the tablet. In cases when such a tablet is used, it will often be useful to apply a threshold to the pen-pressure data, and set the pen-condition flag to “pen up” whenever the pressure falls below the threshold).
The method further comprises segmenting the handwritten symbol into elementary strokes. As will be seen below, this segmentation is typically achieved by identifying natural breakpoints between strokes. Breakpoints are identified by such criteria as abruptness of direction changes, as well as by pen lifts.
The method further comprises recording, for each stroke, a so-called standard parameter set that comprises the x and y coordinates of the stroke's endpoints, the arc length s of the stroke, the net turning angle &phgr; and the relative initial tangent angle &thgr;. The net turning angle is defined with reference to the tangent to the stroke (treating the stroke as a geometrical curve). If the tangent is envisaged as a directed, rigid rod lying against the stroke, then &phgr; is the net angle through which the rod rotates as the point of tangency moves along the curve from the initial to the final endpoint. The relative initial tangent angle &phgr; is defined with reference to the endpoint vector; i.e., the vector from the initial to the final endpoint. More specifically, &thgr; is the angle from the endpoint vector to the tangent at the initial endpoint.
REFERENCES:
patent: 4550438 (1985-10-01), Convis et al.
patent: 4972496 (1990-11-01), Sklarew
patent: 5333209 (1994-07-01), Sinden et al.
patent: 5454046 (1995-09-01), Carman, II
patent: 5559897 (1996-09-01), Brown et al.
patent: 5577135 (1996-11-01), Grajski et al.
patent: 5596656 (1997-01-01), Goldberg
patent: 5610996 (1997-03-01), Eller
patent: 5659633 (1997-08-01), Ilan et al.
patent: 5698822 (1997-12-01), Haneda et al.
patent: 5740273 (1998-04-01), Parthasarathy et al.
patent: 5761328 (1998-06-01), Solberg et al.
patent: 6044174 (2000-03-01), Sinden
patent: 58201183 (1983-11-01), None
Tappert et al. “The State of the Art in Online Handwriting Recognition.” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 12, No. 8, Aug. 1990, pp. 787-808.
Chang Jon
Cohen & Pontani, Lieberman & Pavane
Lucent Technologies - Inc.
LandOfFree
Method and apparatus for parametric representation of... 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 and apparatus for parametric representation of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for parametric representation of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3136405