Establishing a Standard Stroke Database

through Acoustic Analysis

Angelica Joelle P. Bañez, Tifanny A. Fernandez, Vince Lawrence E. Martin1, Ariana
Marie R. Reyes, Viel Ellison V. Tan, Gabriel Rodnei Geslani, M.Sc.2, Harold Alexis Lao
Department of Electronics Engineering, Faculty of Engineering
University of Santo Tomas – Manila, Philippines
1
[email protected], [email protected]

Abstract—Assessment of an individual’s handwriting can be and curvature in relation to the other strokes in a letter [Skills for
valuable in identifying existing or potential medical problems. action]. Some of the basic factors in handwriting include the
However, it cannot be implemented yet because there is no writing’s pressure, form, speed, spacing, size and the degree of
database which can provide the baseline for the dynamics of slant [TL Smith]. To accurately assess one’s penmanship and to
handwriting, thus there is a need for a standard database of such consider the dynamics and basic factors of handwriting,
dynamics. In establishing the standard database of handwriting, handwriting recognition must be done.
three methods are possible, namely through Optical Character
Recognition (OCR), Inertial Measurement Unit (IMU)-based Handwriting recognition has been used in the medical field
pens, and Acoustic-Based recognition. There have been previous for a long time. One method of handwriting recognition is
researches conducted in handwriting recognition which employed through Acoustic Analysis. Through this method, it can provide
the use of the three methods. However, the following methods were distinct characteristics for the dynamics of handwriting, can also
only used for writer identification and results showed that only be able to recognize small handwriting since sound is of utmost
through Acoustic Analysis where the writing instrument importance. This method can also measure the basic factors in
generated sounds provided a rich source of information for writing analysis. Most importantly, the visual quality of
pattern recognition based on the distinct features of the sound handwritten text will not matter and affect the results which will
signal as compared to OCR and IMU-based pens. This paper aims be beneficial for people having poor handwriting. The
to determine the distinct signal features of the dynamics of aforementioned dynamics and basic factors of handwriting play
handwriting after feature extraction such as time, scale, pressure,
a crucial role in identifying if a person exhibits a poor
and stroke through Acoustic Analysis. The data set consists of the
different handwriting strokes which constitutes the lowercase
penmanship. However, one cannot simply diagnose a person’s
English alphabet and participants from the 5th year UST ECE handwriting without having a baseline for it.
students were asked to write the data in order to differentiate the Having a baseline will be useful in determining the
characteristic signals each stroke of a letter. handwriting errors. Establishment of the correct stroke or
handwriting database will make the assessment easier and more
Keywords—handwriting analysis, acoustic analysis, stroke
accurate. Studies have shown that there have been existing
database, support vector machine, k-nearest neighbor
databases with the correct strokes however, existing handwriting
database only include Arabic [J. Rmadam] and Tibetan
[Tibetan], and the complexity of the both languages’ script
I. INTRODUCTION introduces a big difference with the English handwriting
Writing is one of the most essential skill a person can [Tibetan] [Arabic]. This study aims to establish a standard stroke
develop. The development of a person’s writing ability deals database through Acoustic Analysis. Establishing a standard
with his motor control, kinesthesis, also an integration of database for the English alphabet will be beneficial in
cognitive and motivational elements [Laszlo] [reisman]. Early determining the accurate way of writing letters. Having a
assessment of a person’s handwriting may be useful in standard database for the correct handwriting stroke can be
diagnosing if a person exhibits poor handwriting, which may useful for occupational therapists in assessing a person’s
indicate a development coordination disorder or a neurological handwriting.
disorder e.g., dysgraphia [DTW] [Rosenblum]. Evaluating a A. Selecting a Template (Heading 2)
person’s handwriting mainly uses personal judgment as the
primary method [P. Addy]. However, personal judgment is A. 1st few sentences: handwriting, importance of hw,
always not reliable and may not exactly identify the dynamics B. Specific area – database, analysis
of handwriting errors [DH Stott]. The dynamics of handwriting
states that every letter is written by linking two or more strokes C. Introduce specific topic of research and explain
in a sequence, and each stroke has a distinct direction, length, importance / benefits – database thru acoustic

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

 D. Past attempts and difference with research [2] J. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol.
2. Oxford: Clarendon, 1892, pp.68–73.
E. Conclude by mentioning specific objetive”. [3] I. S. Jacobs and C. P. Bean, “Fine particles, thin films and exchange
anisotropy,” in Magnetism, vol. III, G. T. Rado and H. Suhl, Eds. New
York: Academic, 1963, pp. 271–350.
[4] K. Elissa, “Title of paper if known,” unpublished.
II. EXPERIMENTAL
[5] R. Nicole, “Title of paper with only first word capitalized,” J. Name
III. RESULTS AND DISCUSSION Stand. Abbrev., in press.
[6] Y. Yorozu, M. Hirano, K. Oka, and Y. Tagawa, “Electron spectroscopy
Before you begin to format your paper, first write and save studies on magneto-optical media and plastic substrate interface,” IEEE
the content as a separate text file. Complete all content and Transl. J. Magn. Japan, vol. 2, pp. 740–741, August 1987 [Digests 9th
organizational editing before formatting. Please note sections A- Annual Conf. Magnetics Japan, p. 301, 1982].
D below for more information on proofreading, spelling and [7] M. Young, The Technical Writer’s Handbook. Mill Valley, CA:
grammar. University Science, 1989.
IEEE conference templates contain guidance text for
ACKNOWLEDGMENT composing and formatting conference papers. Please
ensure that all template text is removed from your
REFERENCES conference paper prior to submission to the
conference. Failure to remove template text from your
paper may result in your paper not being published.
[1] G. Eason, B. Noble, and I. N. Sneddon, “On certain integrals of Lipschitz-
Hankel type involving products of Bessel functions,” Phil. Trans. Roy.
Soc. London, vol. A247, pp. 529–551, April 1955. (references)